4 Ways to Re-Engage High Value Cannabis Consumers with Andrew Watson at Happy Cabbage Analytics

{}Introduction

Host: ciao everyone, welcome to the Cannabis Marketing Live podcast, where we cover the latest and most effective marketing trends and strategies to grow your cannabis dispensary, delivery service, or brand. I'm your host, and today we're discussing four ways to re-engage high-value cannabis — we're joined today by Andrew Watson, he's the CEO and founder of Happy Cabbage Analytics, welcome to the show, Andrew.

Andrew Watson: thank you for having me.

Host: yeah, of course, of course, before we kick it off today I'd like to give a shout out to our sponsor, MediaJel — MediaJel is the leading marketing platform helping cannabis brands reach consumers through their compliant ad network, with real-time reporting and analytics dashboard, as well as conversion tracking. well, let's kick it off here today, Andrew, tell me a little bit about your background and Happy Cabbage.

Andrew Watson: yeah, so, Happy Cabbage, we make revenue easy for cannabis retailers nationwide, by using the power of data science to take their first party data and turn it into revenue through predictive SMS marketing, inventory, product recommendations, and other tools that really make data and revenue generation easy for cannabis retailers. myself, I have a background in data science, coming from basically tech in San Francisco, and really excited to be able to bring a lot of those technology trends that happen outside of cannabis to cannabis, to really help us all make more money.

Host: really, yeah, I mean retailers need it more than ever, especially now, right, especially those California retailers, so anything we can get, we're gonna take. and you mentioned — was it recommended, or kind of automated text messaging — can you be more specific and give us an overview?

Andrew Watson: yeah, yeah, so what we do is we sit on top of your point of sale system, and we measure how often individual consumers are purchasing and what they're buying, and then we also have a prediction on, hey, if this set of a thousand consumers were to receive a text message today, this percent of them will convert, and therefore you'll make this amount of money, and we have learning models that are basically running every single day and measuring this trend across hundreds of retailers. and so what we can do is we can make it as easy as a click of a button, where we say, if you text these 300 people you will make two thousand dollars, would you like to send a message? right, and so it's very automated segmentation, we do provide you the tools to make it yourself if you want to, but basically we're like, look, the computer can do a much better job at selecting who you should message than a human can, objectively, so we just bring that technology to make it very powerful but very easy, which is really key.

Host: yeah, I remember doing, managing SMS campaigns like back in 2015-16, and I had to spend a ton of time segmenting customers, figuring out who bought what, figuring out what messages to send, so you pretty much cut out all that manual work, and it's all replaced with software and probably AI, right, and then you can do this like forecast modeling and everything?

Andrew Watson: that's nice, exactly.

Host: yeah, do the customers know about this, do they just log in and do this, or do they get an email like, hey, you have this opportunity —

Andrew Watson: yeah, yeah, so we have an interface, it's really simple, you just log in, and the first thing it says is, for example, one — you want to just show us?

Host: oh yeah, it's probably easier — yeah, I could just show you, if you would like to see.

Andrew Watson: so let me just really quickly pull this up.

Host: give us the tour, give us a tour.

Andrew Watson: yeah, yeah, yeah, once again — yeah, I didn't know that you did the forecasting and — yeah, so this is just a simple demo version of it, but what you can see here is what it's doing is the computer is — and it's basing it off of how well the computer did this previously, right, because we have the data on all the previous transactions and how often people buy, and then we also have the data on all the previous messages that have been sent millions of times, we know like what's the intersection, what's the overlap, what's the proper time for someone who's buying every seven days, or somebody's buying every 30 days, for them to receive a message. and then what we also have found is that messaging about what they actually prefer can double the efficacy of the message, and so what we've done here is we basically said, hey, we've automatically identified 340 people, we know they prefer flower, and we have a prediction, and all of this is dynamic to five percent of order again, you'll get seventeen hundred dollars in sales over seven days, you can write the message and you can click send, and then be done, basically. and so a lot of our customers who get the most effective results, the only thing that they do is they just blow through the opportunities and then they go about their day. we do also have — and this demo doesn't have it because we haven't loaded in the example data — but you can create your own custom campaigns, and you can get very specific into demographics and into product preferences and stuff, and what it'll do is it'll detect how well that campaign you came up with did, and then give you a prediction of how well it'll do again. so if you are wanting to do a bunch of your own analysis and build your own campaigns, you know, like, three or more times, eight or more times, you know, at like, buy in the afternoon, and you think that that's very powerful, the computer can say, okay, send it, and then based on how well that did, this is how well it'll do if you did it again.

Host: yeah, that's awesome, yeah, I know we have some customers that, you know, Mondays through Wednesdays are slow days, they want to try to get people in during those days, or through happy hours or something like that, just to even up the traffic flow in the store, because, you know, Thursday through Sunday it's just mayhem in some cases.

Andrew Watson: no, I like that.

Host: yeah, yeah, why should cannabis companies put effort in re-engaging

Why Should You Re-engage Lapsed Customers?

Host: latent or lapsed customers?

Andrew Watson: yeah, so the average retailer we run into, who has about two years of data, 50 — so half of all consumers who are opted in to receiving messaging haven't shopped in the previous 90 days. the average order frequency we see across all of the shops we deal with is approximately 25 days, so most consumers shop every 25 days. when half of your consumers who previously were shopping every 25 days now haven't shopped in the previous 90 days, that can be thousands, if not tens of thousands, of potential consumers who — you know their name, you know their address, you know what they bought, so you know what they like to buy, you know how often they buy, you know how long it's been since they bought, you know what time of day, you know day of week, you know everything about these consumers. and so by always having a strategy — and that's why one of our primary opportunities is "recapture lost customers" — a strategy of going after and bringing in those latent customers back in, basically, it's your most low-hanging fruit.

Host: yeah, right.

Andrew Watson: your cost of reacquiring those customers is minuscule compared to the cost of having to go out and acquire brand new customers off of, you know, services like Weedmaps or billboards or something like that, it's just, we're part of stable states, right.

Host: yeah, exactly, like it's kind of like, you know, you should be doing that, right, like in classic marketing theory, like you definitely should be constantly nurturing those who are falling off.

Andrew Watson: yeah, you should nurture them.

Host: you know, it's much more cost effective to engage with your existing customers, you have all these data points on them, activate that data, figure out a way to get them back in the store, whether it's working with your vendors to create a vendor day and then send people to — everyone that's interested in that brand for that vendor day, you can segment by that, right, get those people in the door, it's just so much money left on the table.

Andrew Watson: yeah, I mean, it's just business one-on-one, like, you have to nurture your existing customers, and within cannabis it's even more important, right, like email and SMS are fantastic, and push notifications as well, and you really just have to stay top of mind, so make sure that you're engaging with them and you're not spamming them, you're not sending the message every day, but make sure it's a curated message that's based on what they have purchased in the past, which is already built into Happy Cabbage, so all that stuff you can cut out, you cut out all the manual work you have to do, choose the opportunities, choose the vendors, brands that you want to highlight, make it work for yourself, and make sure that you re-engage with that 50 percent of people in that 90-day lapse time period, like that's just money on the table, and if you're not re-engaging with your

Re-Engaging Your Customers to Grow Revenue

Andrew Watson: customers, I don't know how you're going to grow your revenue month over month, like you're going to acquire new customers to grow your revenue, but you need to retain your existing customers, you get that compounded growth, so if you don't have that in place, you're just leaving money on the table.

Host: and, you know, I'll just add on, another couple strategies for re-engaging lapsed customers — now obviously at MediaJel we do programmatic advertising, so someone comes to the website, we pixel them, we continue to follow them around the web, anywhere they are, you know, if you think about going to Amazon, you're looking for shoes, you go to Instagram, you start seeing those shoes everywhere, it's the same thing, right, you want to follow your customers around, you can do that on programmatic, you can do that on Google as well. so just keep that in mind when you are figuring out ways to engage with your customers, and, you know, the best way is to engage with them in person, right, so they're in your store, make sure that you are prioritizing your top 20 percent of customers, that's another segment that I —

Andrew Watson: yeah, I feel like cannabis companies aren't leveraging that, like that top 10, 20 percent of your customers, you know, they probably don't take 25 days to shop, right, they're probably coming by every week, so have some type of messaging to make sure that they're coming back in the store, and then offer them loyalty points or some type of benefit to get them coming back, and whether it's vendor days, or what I always like to suggest is have an exclusive strain drop, or product drop, that's only available to the VIPs.

Host: yeah, it's just like — boom — it's a no-brainer, so you have all this first party data that you can leverage, so use it. Andrew, can you define what first party data is for us?

Defining First-Party Data in Cannabis

Andrew Watson: yeah, so first party data is data that, through the nature of your business, you've collected information about your consumers, and this could be if they're visiting your website, you know their IP address, you can know information about if they've logged into your menu, like what their email address is, information about that — if you are leveraging like an online audience or pixel type of stuff, you can normally map that to even more information. and then what's key in cannabis, which is one of the reasons why I started working in this industry as a data scientist, is your point of sale system also has all of this information about every single customer who's come through the door, right, so not only can you leverage all of the modern web technology that exists to be able to understand who's visiting the website, but you also have something that retailers outside of cannabis just don't have, which is every time someone buys something they need to check in, and with that you can connect the entire funnel all the way through, and you have very powerful information on what people want to buy. right, if you think about when you go into a store outside cannabis, every single time you buy something they ask for your email address, then you only give that 20 percent of the time maybe, whereas in cannabis, a hundred percent of the time you give your ID.

Host: so, oh yeah, with that kind of information, right, you have so much more power.

Andrew Watson: and that's why I think first party data in cannabis is about ten times, if not more, more valuable than it is outside of cannabis, because of just how rich and how complete it is.

Host: I agree, yeah, and, you know, there's a reason you go to Walgreens or CVS or Safeway or any store, there's a reason that they want you to join their loyalty program, there's a reason they're giving you discounts, because that data is worth more than oil, right.

Andrew Watson: yes, I don't know about nowadays because oil's gone through the roof, but, you know, I know that that's how Facebook, Amazon, Apple, and all these other companies, Tesla — you know, they're all data companies, right.

Host: yeah, they're — what's the difference between first party and second party data?

Andrew Watson: yeah, so second party data is basically data that you're buying from somebody else, very much, right, like I would say there's another company, right, that — this happens, New Frontier Data or something —

Host: yeah, exactly.

Andrew Watson: like they've mined data, honestly, they've mined other people's first party data, I'll put that out there, and they basically sold it back to you at a massive markup, because probably the people they got it from didn't realize how valuable it was in the first place, that's essentially second party data, and also, same thing, because there's so much rich first party data in cannabis, there happens to be a lot of second party resellers in cannabis as well.

Host: yeah, there's so much data available out there, you know, whether you're — number one, you should be capturing yourself. so we should probably talk a little bit about that right now, like, Andrew, what are some of the best ways you've seen first party data captured by a brand or a retailer?

Andrew Watson: yeah, so for retailers it's actually very simple, I will say, if you've made a good decision on a POS purchase, and those POS systems have very good, complete data — so for example a scanning technology on the ID that'll capture all that information off the ID, and then you have stops in place where you are collecting like phone number, email, other information — oftentimes POSes can even enforce your bud tenders or your receptionist to have to capture that information, which helps a lot, right, I would say that's the first line of attack, so to speak, on data collection, and that's where your largest source of truth data is coming from. on top of that, then your e-commerce system that you have as well, being able to make sure that you're collecting just as much data on that side as you're collecting with the in-store interactions, right, and so that means things like making sure you have an e-com checkout flow, an e-com system that is encouraging logins, that's encouraging people to fill out profiles, but isn't necessarily too onerous, so that you're not blocking potential transactions, right. and then from there, it's supplementary systems that can help add more contextual data on top of that, right, and so that's things like sending out surveys, getting feedback data, e-commerce systems that have reviews and stuff like that, you can actually connect a review to a person, you can get sentiment from that person, you can also do things like VIP programs, loyalty programs, that kind of stuff, can collect some more of that supplementary data. but I really do think that — and I'm saying this as someone who spent a lot of time doing healthcare data at work — the holy grail of data is having a robust connection from e-commerce to POS, where you're collecting all the information all along the way.

Host: yeah, yeah, there's some key information you want to collect about people, right, and

Data Collection and Product Information Management

Host: obviously their first name, last name, email address, phone number, those are great ones to collect, I mean pretty much standard, I know that from the check-in they're — you're automatically going to collect their driver's license information, right, so their address, date of birth, driver's license number, everything like that, and then when I come into the dispensary I check in, I buy — you know for me I love my pre-roll, so I'll get my pre-rolls, I can smoke those on the run, some sleeping edibles, and then I've now recently loved my cannabis drink, so I come in the store, I buy all those, you know, next week I can expect a message from Happy Cabbage or the retailer that, hey, Ken is having a vendor day here on Friday, and they're giving out 50 percent discounts on their drinks, he should come because he bought drinks in the past.

Andrew Watson: it's like — yeah, there you go, yeah, a hundred percent, yeah. one thing I'll say there as well, one other really important thing about first party data capture, is that it's awesome if you have all the information about the consumer, but what's even cooler is if you have all the information about the consumer and then all the information about every product that they've ever bought as well. so making sure that you are using a POS, or that you are at least curating and controlling how you're cataloging product in your system — you know, you mentioned Ken, right, making sure that you're using the brand features in your POS, making sure that you're appropriately using SKUs, product names, that kind of stuff on your e-commerce menu, connecting that through to your POS correctly, that's gonna give you a lot of rich information as well, so that categorically, hey, here are the Ken lovers, here are the beverage lovers, here are the sleepy edibles lovers, versus needing to have all this rich consumer data but then not knowing anything about anything they've ever bought, right, and that's also super key and super important.

Host: yeah, it's you know, when I go into a store that I have already shopped at, I expect them to know what I want, like I'm going in, I don't want to go to the dab area, it's not my deal, I want to see the flower, I want to see some of the edibles, and that's it, and if more advanced retailers, they'll typically have a bud tender with an iPad, and they'll say, oh, you know, based on what you purchased in the past, is what I think you'd like, and then take you in that direction, right.

Andrew Watson: that's what you'd expect, I mean that's what you get from, I would say, more advanced experiences within retail stores.

Host: right, yeah, and that's — extra, yeah.

Andrew Watson: yeah, we, because we have that data on the SMS side, we can profile the consumer and then figure out, hey, this is exactly the message to send to this consumer based on what they bought in the past, we also have an in-store recommendation tool as well that we sell with it, so it comes along with it, and that exactly, like a retailer can use that, look up the consumer on the iPad, and based on the real-time inventory at that store and that person's previous purchases, it'll actually give you a recommendation on exactly what to sell that person, that is going to optimize the price point, and we've actually seen people who use this tool to provide that recommendation based on your previous purchases see a 13 percent uplifted ticket size, because people are buying and selling things at full price, because they're more attuned to their preferences versus driving people to whatever product is discounted at this point in time. so again, other ways you can use first party data to augment the in-store interactions, don't just use it to market, but then, like, the full funnel is all the way from discovery of customer to that person making a purchase, right, and so make sure you're mapping and using the data all the way through.

Host: yeah, exactly, and, you know, you're collecting all this first-party data, Andrew, how can cannabis

How Should Cannabis Retailers Sort Their First-Party Data?

Host: retailers sort these audience segments and monetize them?

Andrew Watson: yeah, so that's where you can spend a lot of time sifting through this information, you can spend a lot of time doing analysis paralysis, and at the end of the day come up with a set of segments that you may believe exist, but perhaps don't have that much actual power when you go out and put them out into the world, right. and so what I always encourage is thinking of things based on recency, frequency, latency, that kind of stuff — how often is someone buying, how long has it been since they last purchased, that's going to be one of your largest driving factors when it comes to figuring out how much you should expect an audience to convert if they receive a message, or whether or not they need to be receiving messages or content, right, and so that could be your people who have fallen off, your 90-day shoppers, your weekly shoppers, your monthly shoppers, your bi-weekly shoppers, that kind of stuff. then from there I also heavily encourage thinking of product-category-based segments as well — what we have seen is that if you message somebody about a product category that they have purchased in the past, you will have twice — 2x, not 20 percent, not 30 percent, 200 percent — increase in the efficacy of that campaign, right. now, people who only buy edibles do not want to see your dab deal, it's just things they do not care about, right, I mean that leads to unsubscribes, it leads to low-to-no conversion, and then also conversion rates by category are very different — vapes, for example, we see very low conversion rates overall when people receive vape-based deals, for people who have vape preferences, dabs — people who buy dabs, a lot of primary dab preferences, who receive dab-based content, have significantly higher conversion rates than any other segment that we go after. and so I think thinking of things — how often people are buying and what they're buying — is really important, particularly in cannabis, when you think

Factors to Consider in Cannabis Consumer Behavior and Engagement

Andrew Watson: of the range of potency and the range of modality of products, is very very high, so you don't want to just do one-size-fits-all, because you're leaving potentially thousands, if not tens of thousands, on the table. and then, finally, I do encourage thinking of things like how far are people on their journey with knowing your brand and knowing your experience, right, and so that could be, like, who are your first-timers who have just tried you out and potentially are falling off, so you need to encourage them to use a delivery channel, for example, because they're far away, right, who are your high-frequent customers who for some reason keep coming back, and understanding their mindset and trying to think of ways that you can communicate to them in ways that are effective and that matter to the way that they're purchasing — you know, "we miss you," "happy Valentine's Day," "it's been a while," like, your birthday's an easy one, right.

Host: like birthday, that's a great one, that's a great one.

Andrew Watson: demographics, right, so thinking of, you know, we do see a lot of difference when it comes to seniors versus your Gen Z, particularly if your dispensary's close to a college, I will say, and so understanding that mix and types of things can help out a lot.

Host: it's good to know, is there any — I mean you highlight some campaigns, is there any that are your favorite, that are like, you know, people must include on theirs, I know

Effective Cannabis Marketing Campaigns and Strategies

Host: it's maybe perhaps like an advocacy or a deal campaign, like, is there any that really stick out to you?

Andrew Watson: yeah, yeah, some of the most effective ones that I've seen — hype brands, so brands that have been coming up a lot in your store, re-engaging people about those brands can heavily influence bringing people more back into the store, particularly with those discounts — now I don't necessarily encourage over-discounting on promotional stuff, but the fact of the matter is, you know, Jeeter is a good example, when you message Jeeter customers about the fact that there's a Jeeter deal, their Jeeter sales can quintuple in a day easily, right, and understand that with every one Jeeter product that they're buying, particularly if you have something like a delivery minimum, or while they're already in the store, they're also going to be adding on other products as well, right, and so the retailer on net is going to be making more from that than the brand, right, that's low-hanging fruit — so your top brands, making sure you're continuously communicating about your top brands is low-hanging fruit. and not to disparage more of the niche brands, but unfortunately the data does show that if you are messaging about niche brands at the expense of messaging about top brands, you may be leaving money on the table, and that's just the reality of the situation, because people are gonna be like, I don't know what this product is, they may be unsubscribing, they may get messaging fatigue. so that's one thing, and then the other thing I would say is regular nurture campaigns for lost customers by category are very effective, set it and forget it — make sure your edibles customers who haven't been back in the store in months are constantly receiving some messaging reminding them about edible deals, make sure your flower customers who haven't been back in months are constantly receiving some messaging about flower, those are the consistent things that each individual campaign may not get you a lot of money, but over the course of a month you can be bringing in a lot of revenue through them.

Host: yeah, a hundred percent, I know one that we love to do, and it's not really re-engaging, but it's more like a store launch campaign, we'll send people to a landing page before the store is open, let's say a month before, get them to join a VIP program, maybe offer like a 20 percent discount on the opening, build a list of 500 to 1,000 people, and then, on the day before grand opening, send out that message, and, you know, you have a line around the corner, and the owners of the dispensary — if it's a chain — they're gonna be happy, and it's not even the revenue, but it's the perception that the neighborhood and everyone around sees that you're a high demand retail store, right, and that people are standing in line for you, it's the same thing when you're standing outside a nightclub or something, right, people get FOMO.

Andrew Watson: yeah, they want to go check it out.

Host: yeah, word of mouth is an incredibly effective way to drive distance.

Andrew Watson: incredibly.

Host: yeah, yeah, exactly, and then, you know, I didn't mention it before, is combining SMS, email, and programmatic advertising to do more of an omnichannel marketing approach, and also, you know, SEO, to capture the keyword and search behavior — those are great ways for us to leverage this, and, you know, given the integration that we have between our two companies, I think it's

Display Ad Case Study: Impressive Results from Targeting with First-Party Data

Host: probably time we should show a case study here from one of the campaigns that we launched together, right?

Andrew Watson: yeah.

Host: so let's just look at this — so these numbers are impressive, let's just say the least. so we were in a test campaign, using the first party data from Happy Cabbage, and then we activated that data through MediaJel to reach these cannabis consumers through mainstream publishers, right, like your ESPN, your, the Chive, TMZ, Sports Illustrated, GQ, Salon.com, like all the mainstream publishers that you would expect. so Mankind spent 91 dollars on this campaign, so that is paying for placement on these publications, and they were able to generate 15,000 dollars in revenue from that, so I don't even know what the return on that is, but it's, it's very large, very large. so it's really an effective way for you to get in front of people, to engage with them, you know, you can send a text message, you can send out an email, maybe they don't respond, maybe they're distracted, maybe they're too high that day, whatever it might be, if that's the case, then you still want them to see ABC Dispensary, when they're playing Words With Friends, when they're trying to swipe right on a dating app, trying to find their next match. so a lot of opportunity there to get more awareness and just stay top of mind, really — it's the marketing rule of seven, the old-school way to say people need to see your brand seven times before purchasing, I think it's probably higher than that now, given all the attention economy and all the distractions that are available to us through our mobile phones nowadays, so something to consider. and then, looking at this campaign, we can actually see the lift from all the products, so you can look at the individual products, and you can see — oh, okay, well, I see Berry Fruit, True Fruit, that one's getting a lift for their 10-packs, Yuzu Lemon, here's another top product that's been effective in our marketing campaigns here. so, getting — oh, here we go, I just sorted by quantity, so it looks like Zen GMO, for this particular campaign, sold 20 units, Sativa 10-pack, Peach OZ, they're all — these are units that you maybe would have moved if you weren't running a programmatic display advertising campaign — my favorite strain of all time, Durban Poison, right there, number eight, eight sales. so just keeping that in mind, and you can also leverage this data to run a co-branded, co-marketing campaign, so let's say

SMS Case Study: Retargeting Campaign Powered by Customer Data

Host: Jeeter wants to move units at ABC Dispensary, they could run ads with Jeeter, I'll actually show you one — who am I running with, Jeeter's Green — I don't know if this one's active now, but you could showcase a brand as the actual advertisement, and work out a deal, or maybe you get a price break on purchasing, or maybe they give you some money to invest in marketing, so we're all in this together, and, you know, brands want to see the products sell through in a retail store, and retail stores want to get that product off the shelf as much as possible, so keep that in mind.

Andrew Watson: yeah, and that's what was really cool about this campaign, and I always like to say that SMS is just a channel, right, there are many different channels, this is not the be-all and end-all of channels, and what we've seen through usage of SMS is that the determination of efficacy is whether or not you leverage first party data to match people to when it's proper for them to be receiving content, and proper for them to be converting, and what content that is. what's really cool about this campaign is what we did, is we used the same machine learning technique that we use to identify who are the proper people to text, and we basically took that, applied it to the entire data set of the client, and then handed those device IDs over to MediaJel, and they were able to run this really awesome campaign against that data set, and get really great first party data back on individual transactions, right, and so that's why we were able to take 90 dollars and turn it into 15,000, exactly. but even within that, we got data back based on this that can help us retarget even further, right, and help lift that even further, and that's the beautiful thing about data science and linking the channels together, is you can make it learn, and you can make it grow, and make it become more effective over time, and to that point, if you have brand partners also engaged on this, right, being able to have a machine select out the individuals who are most primed to purchase Jeeter, who are falling off of that, feeding that selection of consumers back into the machine over and over and over again, so that those people are constantly receiving advertisements that are branded, that are bringing them back to your store to purchase that targeted brand, and we know, because again this like giant feedback loop, how to make that more effective over time, and so that's when we can take 90 and start turning it into 20, 30, 50 thousand, right, now I don't want to over-promise on anything, but

Host: yeah, like, yeah, I mean it's there, and you know, just looking at, we shoot for a 0.2 click-through rate on traditional campaigns, so by leveraging first-party data, it looks like we're about six times, seven times more effective at getting people to click on an ad. all right, so you're not wasting a bunch of money on showing ads — wait, I guess not waste, but you're not spending a bunch of money on people who are not your existing consumer, or not your consumer, you are advertising to people that have already purchased with you, and

Maximize Your Ad Spend with First-Party Data

Host: the way that we connect what you're doing, Andrew, on your side, to what we do, is an email address, right, an email address, mobile advertising ID as well, those are the two ways that we can connect, that's essentially like a social security number for your cell phone, right, so we can target ads based on that, and you can see here is a sample of, "20 off your next cannabis delivery," so this was, you know, you're trying to re-engage them to purchase again, looks like they probably lapsed for a while here, and then you can see the coverage, right, so, you know, from my experience, the top categories that are effective from

Find Opportunities to Re-engage Cannabis Customers On Their Favorite Media

Host: programmatic advertising is games, right, so everyone has games on their phone, no one likes to pay for them, so the way that these games monetize is through ads, all right, so games are huge, Reddit is on here, local — CBS, CBS San Diego, this is for Mankind in San Diego — podcasts, streaming, right, weather app, OfferUp, you know, it's basically — there's social media here, a little, memes, meme sites do really well, social media sites do really well, dating sites — Grindr is one of the highest performing publishers we work with, that category does really well, any dating, and especially gay dating does really well, podcast players — so you can just get an overview of the coverage, and at MediaJel we have 75,000 publishers in this, so there's quite a lot of opportunities for you to get in front of cannabis consumers, and you know that they're a cannabis consumer because we got that data from Happy Cabbage, and then we make sure that we distribute that data to all these publishers, which is all the publishers that accept advertising, and then looking at how they actually engage with you — so you can see, I'm going to change this to a five-day attribution window, so you can see that it ranges — some people only need to see you four times before they'll come by again, this person's eight times, if I open up their attribution window to all time, looks like some of these customers have to see us 50 times to come by again, all right. so the number of times the impression — anytime that they've seen it, right, someone saw this ad on 5/15, looks like they saw it four times on 5/15 on their Android phone, on their local breaking news app, and then they purchased on 5/19 for 42.37, and we know this because of an IP address match, so they come by again, and then now we're gathering more first party data about them and enriching the data set that we already have. so just a lot of value here to really maximize your return, this is a 419-to-one return on your investment, so it's absurd, to say the least.

Andrew Watson: well, yeah, there's just a lot of opportunity there to generate revenue for your retailer, I would say it's — the results are so compelling that if you're spending money on any retention SMS, retention email, retention loyalty, it's a no-brainer to also be spending money on retention-based programmatic, it's just a no-brainer, right, like, you're jumping all this money into discounting to try to bring people back, you might as well spend it on what the — like, the low-hanging fruit is, right.

Host: exactly, exactly, I mean, companies like Mercedes, BMW, they spend like 50 percent of their marketing budget on existing customers, right, so they have a specific brand positioning, right, luxury vehicles, and people want to associate or relate to and live that lifestyle, so there are reasons that they advertise in the ways that they do, it's, people want to feel part of the community, right, you know, the Tesla people are notorious, like if someone has a Tesla, they're part of the mission, the vision, right, so let's keep that in mind — what are — get me, Happy Cabbage campaigns, Andrew?

Andrew Watson: I'm Happy Cabbage campaigns, so, did I — did I catch you off guard there? I got you off guard — can you — could you be more specific?

Host: yeah, yeah, we can skip that one, we can skip that one. what about — I guess let's take a little break here, because I want to make sure that we answer any questions from the audience — so Ravi, Donald, Colleen, Zach, I don't know, do you have any questions for Andrew that you'd like us to answer live here? let me actually check social media too, because we may have some questions that people are asking on social, but feel free to ask your questions in the Q&A, which is at the bottom of the Zoom interface. just make sure there's no other questions here on LinkedIn — and nothing yet. all right, well, we'll give you some time, if you have any questions post them in the Q&A. all right, so, you want to talk a little bit about maybe the KPIs, right, which KPIs should a cannabis company be

Essential KPIs for Cannabis Marketing Success

Host: tracking, in order to know when to send out a re-engagement campaign?

Andrew Watson: yeah, so I'll actually start a little bit backwards in that — what I think, in terms of KPIs, right, what I think can sometimes get lost from folks is, at the end of the day, you're spending money on marketing to make money, but at the end of the day you are spending money in order to drive revenue, particularly when it comes to the bottom of the funnel marketing, which, you know, SMS campaigns that are very bottom level, right, discounts that you're giving in store, very bottom of the funnel, the primary KPI which you should be obsessed with is ROAS at the end of the day, it's how much money in revenue did you get from a dollar you spent on this campaign, right, particularly given how much low-hanging fruit — you know, these are people who already know your brand, they've already had experience, oftentimes they've had dozens of experiences beforehand, right, it's just a matter of really understanding, am I spending ten dollars, am I getting back a hundred, a thousand, ten thousand, exactly what those numbers are, right. now from there, in order to understand how you can increase or decrease ROAS, I think really strong, important KPIs to understand are conversion rate, right, so what percent of people, of a campaign, came back into a store within 24 hours, within seven days — we also will show recapture rates, things like that, so of all the people you sent out the campaign to, how many of them were latent and came back in, how many of them hadn't been back in the last 90 days and came back in, and how many dollars were associated with that, right. now, in terms of KPIs to figure out when the correct time and moment to send a campaign is, you actually take your results and you optimize your results, and so, for example, we have a tool that lets you see, based on time of day and day of week, what the appropriate time is, right — so let's say you've sent out a few re-engagement campaigns over the course of a week, you look back at that the next week and you say, okay, Saturday afternoon was by far the most effective time to send that campaign, let's resend on Saturday afternoon, let's double down on Saturday afternoons, right. understanding timing of brand deals, understanding timing of holidays, right, you don't want to burn through your list a week before Labor Day, when Labor Day categorically you've seen a certain amount of lift, and you want to double down on that lift — looking at your last year's data for your days leading up to 4/20, to understand how much of your sales in April came from the days before 4/20, things like that, right, I think are really important to understand, so that you're not just pressing the button constantly to send a whole bunch of messages constantly, because what you're just going to end up doing is burning your list, you're going to get some results, don't get me wrong, but they could be 10 percent of the results that you could potentially be getting.

Host: essentially, yeah, makes sense, makes — that same, make sure that you are dedicating a lot of time to re-engaging with your customers and making sure that they're coming back and that you retain them. Andrew, how do you calculate your customer retention rate?

How Do You Calculate Your Customer Retention Rate?

Andrew Watson: yeah, so — great question — yeah, so we think of, when we calculate retention, we think of it as — and, like I said, we use this 90-day mark, we came up with that through a statistical analysis of just, like, frequency and latency, and it's like a generic thing, but the number on retention I'd like to give people is, of customers who are acquired over a time period, what percent of them are still shopping 90 days later, or the converse of them, what percent of them are 90-day lost after that point in time, right. so if you want to say, okay, what is my retention of my consumers that are acquired in April, right, based on how often they're purchasing, by July you actually have a very good sense of what percent of your April consumers are still remaining consumers at your store, right, and then, basically, that's a rolling average that you can track to understand if you're getting better at retaining, or worse at retaining, the customers, right. and then, obviously, churn is just the converse of that, right, it's basically you either could be measuring what percent you're losing, or you could be measuring what percent you're retaining, they're the same concepts, right, just — it's a very important number to know, because people oftentimes measure how many new customers they're getting, but they're not checking how many of those new customers are staying, which is — yeah, not that good of a practice, because, like I said, you're spending so much money acquiring the new ones, you really want to make sure that you're keeping them.

Host: yeah, exactly, and I know that you put together a data brief, the campaign that we ran together, versus some of the other channels, do you want to share your screen and kind of

Case Study: Marketing Campaigns In-Store vs. Website vs. Weedmaps

Host: walk through that, about in-store versus website versus Weedmaps?

Andrew Watson: yeah, let me just open this up right now, and then let me just share my screen here, so you can see my screen, okay.

Host: I can — yeah, you can, yeah, I can see it.

Andrew Watson: yeah, so what we did is we basically looked at — so the point of that campaign with MediaJel was to go after previous customers who had fallen off, bring them back into the store. now what we looked at is how the customers who were reacquired through that campaign compared to customers who were reacquired from the website, in-store, or through Weedmaps, because they also have the integrated Weedmaps ordering, and what we found is that, when the customers are reacquired — we saw a pretty large uplift in average order value, basically, ticket size, a lot of people call it ticket size, right, where the campaign that we ran with MediaJel, people who are coming back into the store were spending about 122 on average, whereas people who are coming back — previous customers who then came back through Weedmaps, reordered through Weedmaps, where they were previously 90 days lost, spent about 95 on average, right. when we look at that a little bit deeper, right, and so we were just talking about retention, essentially what percent of those people, who it's been 90 days since they purchased, come back into the store, what percent of them are still purchasing within 90 days, right, because you want to think about the retention of the people you've recaptured, and what we saw is that the campaign that we ran, which was very targeted to individuals based on the first party data, 78 percent of those stayed purchasing at the store, right, versus 61 percent of those on Weedmaps. the other really important thing is that, by the way the campaign is designed, we're not double-dipping that much, right, the question is, are you basically inventing a new channel that is owned by, honestly, somebody else, that your consumers are constantly now reordering through — whereas with Weedmaps, what you see is that the customers who were recaptured through Weedmaps are reordering through Weedmaps, which means that you need to maintain a pay-to-play status in order to keep those customers who are already yours, whereas with this we were able to recapture them, bring more of them back into the store, and then we're not necessarily double-dipping, right, because they're coming back directly through in-store, through website, right, and then this is just — you can actually see how much of an uplift overall that is. and then, now, in terms of how long, and then how frequent it's been, what you can see is that, again, the online campaign we ran was very targeted, looking at people who are high-frequent shoppers who have dropped off on that 90-day window, and then bringing them back in, so we're not going hundreds of days without them being customers, right, we're able to essentially have the safety net, the safety net that as customers are falling out that bucket, or capturing them, putting them back in — which is why on average it was 93 days from basically seeing the ad to coming back in, their purchase, their most recent purchase, and then their repurchase, and then they were repurchasing after that every 12 days, whereas with Weedmaps you can see that it was 160 days, because it's not very targeted, right, it was a very long time, and when these people came back in, yes, they were more frequent, like they were purchasing basically once every week, but they were purchasing once every week through Weedmaps, basically Weedmaps owns the channel and you are kind of beholden to pay rent on that channel, right, whereas this is also going after people online, but it's incredibly more effective, and incredibly more tailored to the goals of the retailer, as opposed to saying, hey, rent a channel to maintain this way of customers ordering, and you have to basically pay dividends every month — instead, use your own data that you've already captured to bring them back into the store, and then maintain them through your own property, right, in which you're gonna get a higher average ticket size, you're gonna get more people coming back in faster, and then, overall, make more money.

Host: yeah, which is what really matters, right.

Andrew Watson: yeah.

Host: you want to own your customer data, you want to get them to purchase through your channels that you own, and you want to maximize the return, right, well, that's always the key takeaways that I got from that case study.

Andrew Watson: wonderful, I mean, thank you so much for taking time today, do you want to share with everyone where they can reach out to you and learn more about Happy Cabbage Analytics? actually, I'll pull up the website now and post it in the Zoom chat.

Closing Remarks

Andrew Watson: yeah, so while you do that, it's happycabbage.io, you can fill out a contact us form, there's also a chatbot, you can reach out directly, you can also email directly at sales@happycabbage.io if you're interested in getting in touch. we're really excited about rolling out with more partners, these online recapture campaigns with our MediaJel integration, 90 turns into over ten thousand dollars, really excited to bring those returns to more of the industry as well, and we're also, like I mentioned, rolling out soon more things on the inventory side, more things to round out activation data.

Host: okay, so, yeah, wonderful, well, thank you again for joining us today, we really enjoy working with you, obviously you're a great partner of ours, we appreciate the integration and the trust in us to really activate that data. just as a recap for everyone listening in here, this is the Cannabis Marketing Live podcast, we cover marketing trends and strategies for growing your cannabis business, that's programmatic advertising, SEO, paid search, SMS, email, push notifications, e-commerce, like pretty much anything and everything related to cannabis and tech. the next podcast will be airing live next Thursday, September 22nd, at 11 a.m. Pacific, 2 p.m. Eastern, and we'll be talking about bud tenders and how to not let them tank your cannabis brand, and that's going to be with Luna Stower, she's the chief impact officer at Inspired Hardware out of Los Angeles, so be sure to catch us next week for that, and we'll see you then. have a wonderful rest of your week, and we'll get you on the flip side.

Andrew Watson: thank you so much, of course, thanks.

‍

{}Introduction

Host: ciao everyone, welcome to the Cannabis Marketing Live podcast, where we cover the latest and most effective marketing trends and strategies to grow your cannabis dispensary, delivery service, or brand. I'm your host, and today we're discussing four ways to re-engage high-value cannabis — we're joined today by Andrew Watson, he's the CEO and founder of Happy Cabbage Analytics, welcome to the show, Andrew.

Andrew Watson: thank you for having me.

Host: yeah, of course, of course, before we kick it off today I'd like to give a shout out to our sponsor, MediaJel — MediaJel is the leading marketing platform helping cannabis brands reach consumers through their compliant ad network, with real-time reporting and analytics dashboard, as well as conversion tracking. well, let's kick it off here today, Andrew, tell me a little bit about your background and Happy Cabbage.

Andrew Watson: yeah, so, Happy Cabbage, we make revenue easy for cannabis retailers nationwide, by using the power of data science to take their first party data and turn it into revenue through predictive SMS marketing, inventory, product recommendations, and other tools that really make data and revenue generation easy for cannabis retailers. myself, I have a background in data science, coming from basically tech in San Francisco, and really excited to be able to bring a lot of those technology trends that happen outside of cannabis to cannabis, to really help us all make more money.

Host: really, yeah, I mean retailers need it more than ever, especially now, right, especially those California retailers, so anything we can get, we're gonna take. and you mentioned — was it recommended, or kind of automated text messaging — can you be more specific and give us an overview?

Andrew Watson: yeah, yeah, so what we do is we sit on top of your point of sale system, and we measure how often individual consumers are purchasing and what they're buying, and then we also have a prediction on, hey, if this set of a thousand consumers were to receive a text message today, this percent of them will convert, and therefore you'll make this amount of money, and we have learning models that are basically running every single day and measuring this trend across hundreds of retailers. and so what we can do is we can make it as easy as a click of a button, where we say, if you text these 300 people you will make two thousand dollars, would you like to send a message? right, and so it's very automated segmentation, we do provide you the tools to make it yourself if you want to, but basically we're like, look, the computer can do a much better job at selecting who you should message than a human can, objectively, so we just bring that technology to make it very powerful but very easy, which is really key.

Host: yeah, I remember doing, managing SMS campaigns like back in 2015-16, and I had to spend a ton of time segmenting customers, figuring out who bought what, figuring out what messages to send, so you pretty much cut out all that manual work, and it's all replaced with software and probably AI, right, and then you can do this like forecast modeling and everything?

Andrew Watson: that's nice, exactly.

Host: yeah, do the customers know about this, do they just log in and do this, or do they get an email like, hey, you have this opportunity —

Andrew Watson: yeah, yeah, so we have an interface, it's really simple, you just log in, and the first thing it says is, for example, one — you want to just show us?

Host: oh yeah, it's probably easier — yeah, I could just show you, if you would like to see.

Andrew Watson: so let me just really quickly pull this up.

Host: give us the tour, give us a tour.

Andrew Watson: yeah, yeah, yeah, once again — yeah, I didn't know that you did the forecasting and — yeah, so this is just a simple demo version of it, but what you can see here is what it's doing is the computer is — and it's basing it off of how well the computer did this previously, right, because we have the data on all the previous transactions and how often people buy, and then we also have the data on all the previous messages that have been sent millions of times, we know like what's the intersection, what's the overlap, what's the proper time for someone who's buying every seven days, or somebody's buying every 30 days, for them to receive a message. and then what we also have found is that messaging about what they actually prefer can double the efficacy of the message, and so what we've done here is we basically said, hey, we've automatically identified 340 people, we know they prefer flower, and we have a prediction, and all of this is dynamic to five percent of order again, you'll get seventeen hundred dollars in sales over seven days, you can write the message and you can click send, and then be done, basically. and so a lot of our customers who get the most effective results, the only thing that they do is they just blow through the opportunities and then they go about their day. we do also have — and this demo doesn't have it because we haven't loaded in the example data — but you can create your own custom campaigns, and you can get very specific into demographics and into product preferences and stuff, and what it'll do is it'll detect how well that campaign you came up with did, and then give you a prediction of how well it'll do again. so if you are wanting to do a bunch of your own analysis and build your own campaigns, you know, like, three or more times, eight or more times, you know, at like, buy in the afternoon, and you think that that's very powerful, the computer can say, okay, send it, and then based on how well that did, this is how well it'll do if you did it again.

Host: yeah, that's awesome, yeah, I know we have some customers that, you know, Mondays through Wednesdays are slow days, they want to try to get people in during those days, or through happy hours or something like that, just to even up the traffic flow in the store, because, you know, Thursday through Sunday it's just mayhem in some cases.

Andrew Watson: no, I like that.

Host: yeah, yeah, why should cannabis companies put effort in re-engaging

Why Should You Re-engage Lapsed Customers?

Host: latent or lapsed customers?

Andrew Watson: yeah, so the average retailer we run into, who has about two years of data, 50 — so half of all consumers who are opted in to receiving messaging haven't shopped in the previous 90 days. the average order frequency we see across all of the shops we deal with is approximately 25 days, so most consumers shop every 25 days. when half of your consumers who previously were shopping every 25 days now haven't shopped in the previous 90 days, that can be thousands, if not tens of thousands, of potential consumers who — you know their name, you know their address, you know what they bought, so you know what they like to buy, you know how often they buy, you know how long it's been since they bought, you know what time of day, you know day of week, you know everything about these consumers. and so by always having a strategy — and that's why one of our primary opportunities is "recapture lost customers" — a strategy of going after and bringing in those latent customers back in, basically, it's your most low-hanging fruit.

Host: yeah, right.

Andrew Watson: your cost of reacquiring those customers is minuscule compared to the cost of having to go out and acquire brand new customers off of, you know, services like Weedmaps or billboards or something like that, it's just, we're part of stable states, right.

Host: yeah, exactly, like it's kind of like, you know, you should be doing that, right, like in classic marketing theory, like you definitely should be constantly nurturing those who are falling off.

Andrew Watson: yeah, you should nurture them.

Host: you know, it's much more cost effective to engage with your existing customers, you have all these data points on them, activate that data, figure out a way to get them back in the store, whether it's working with your vendors to create a vendor day and then send people to — everyone that's interested in that brand for that vendor day, you can segment by that, right, get those people in the door, it's just so much money left on the table.

Andrew Watson: yeah, I mean, it's just business one-on-one, like, you have to nurture your existing customers, and within cannabis it's even more important, right, like email and SMS are fantastic, and push notifications as well, and you really just have to stay top of mind, so make sure that you're engaging with them and you're not spamming them, you're not sending the message every day, but make sure it's a curated message that's based on what they have purchased in the past, which is already built into Happy Cabbage, so all that stuff you can cut out, you cut out all the manual work you have to do, choose the opportunities, choose the vendors, brands that you want to highlight, make it work for yourself, and make sure that you re-engage with that 50 percent of people in that 90-day lapse time period, like that's just money on the table, and if you're not re-engaging with your

Re-Engaging Your Customers to Grow Revenue

Andrew Watson: customers, I don't know how you're going to grow your revenue month over month, like you're going to acquire new customers to grow your revenue, but you need to retain your existing customers, you get that compounded growth, so if you don't have that in place, you're just leaving money on the table.

Host: and, you know, I'll just add on, another couple strategies for re-engaging lapsed customers — now obviously at MediaJel we do programmatic advertising, so someone comes to the website, we pixel them, we continue to follow them around the web, anywhere they are, you know, if you think about going to Amazon, you're looking for shoes, you go to Instagram, you start seeing those shoes everywhere, it's the same thing, right, you want to follow your customers around, you can do that on programmatic, you can do that on Google as well. so just keep that in mind when you are figuring out ways to engage with your customers, and, you know, the best way is to engage with them in person, right, so they're in your store, make sure that you are prioritizing your top 20 percent of customers, that's another segment that I —

Andrew Watson: yeah, I feel like cannabis companies aren't leveraging that, like that top 10, 20 percent of your customers, you know, they probably don't take 25 days to shop, right, they're probably coming by every week, so have some type of messaging to make sure that they're coming back in the store, and then offer them loyalty points or some type of benefit to get them coming back, and whether it's vendor days, or what I always like to suggest is have an exclusive strain drop, or product drop, that's only available to the VIPs.

Host: yeah, it's just like — boom — it's a no-brainer, so you have all this first party data that you can leverage, so use it. Andrew, can you define what first party data is for us?

Defining First-Party Data in Cannabis

Andrew Watson: yeah, so first party data is data that, through the nature of your business, you've collected information about your consumers, and this could be if they're visiting your website, you know their IP address, you can know information about if they've logged into your menu, like what their email address is, information about that — if you are leveraging like an online audience or pixel type of stuff, you can normally map that to even more information. and then what's key in cannabis, which is one of the reasons why I started working in this industry as a data scientist, is your point of sale system also has all of this information about every single customer who's come through the door, right, so not only can you leverage all of the modern web technology that exists to be able to understand who's visiting the website, but you also have something that retailers outside of cannabis just don't have, which is every time someone buys something they need to check in, and with that you can connect the entire funnel all the way through, and you have very powerful information on what people want to buy. right, if you think about when you go into a store outside cannabis, every single time you buy something they ask for your email address, then you only give that 20 percent of the time maybe, whereas in cannabis, a hundred percent of the time you give your ID.

Host: so, oh yeah, with that kind of information, right, you have so much more power.

Andrew Watson: and that's why I think first party data in cannabis is about ten times, if not more, more valuable than it is outside of cannabis, because of just how rich and how complete it is.

Host: I agree, yeah, and, you know, there's a reason you go to Walgreens or CVS or Safeway or any store, there's a reason that they want you to join their loyalty program, there's a reason they're giving you discounts, because that data is worth more than oil, right.

Andrew Watson: yes, I don't know about nowadays because oil's gone through the roof, but, you know, I know that that's how Facebook, Amazon, Apple, and all these other companies, Tesla — you know, they're all data companies, right.

Host: yeah, they're — what's the difference between first party and second party data?

Andrew Watson: yeah, so second party data is basically data that you're buying from somebody else, very much, right, like I would say there's another company, right, that — this happens, New Frontier Data or something —

Host: yeah, exactly.

Andrew Watson: like they've mined data, honestly, they've mined other people's first party data, I'll put that out there, and they basically sold it back to you at a massive markup, because probably the people they got it from didn't realize how valuable it was in the first place, that's essentially second party data, and also, same thing, because there's so much rich first party data in cannabis, there happens to be a lot of second party resellers in cannabis as well.

Host: yeah, there's so much data available out there, you know, whether you're — number one, you should be capturing yourself. so we should probably talk a little bit about that right now, like, Andrew, what are some of the best ways you've seen first party data captured by a brand or a retailer?

Andrew Watson: yeah, so for retailers it's actually very simple, I will say, if you've made a good decision on a POS purchase, and those POS systems have very good, complete data — so for example a scanning technology on the ID that'll capture all that information off the ID, and then you have stops in place where you are collecting like phone number, email, other information — oftentimes POSes can even enforce your bud tenders or your receptionist to have to capture that information, which helps a lot, right, I would say that's the first line of attack, so to speak, on data collection, and that's where your largest source of truth data is coming from. on top of that, then your e-commerce system that you have as well, being able to make sure that you're collecting just as much data on that side as you're collecting with the in-store interactions, right, and so that means things like making sure you have an e-com checkout flow, an e-com system that is encouraging logins, that's encouraging people to fill out profiles, but isn't necessarily too onerous, so that you're not blocking potential transactions, right. and then from there, it's supplementary systems that can help add more contextual data on top of that, right, and so that's things like sending out surveys, getting feedback data, e-commerce systems that have reviews and stuff like that, you can actually connect a review to a person, you can get sentiment from that person, you can also do things like VIP programs, loyalty programs, that kind of stuff, can collect some more of that supplementary data. but I really do think that — and I'm saying this as someone who spent a lot of time doing healthcare data at work — the holy grail of data is having a robust connection from e-commerce to POS, where you're collecting all the information all along the way.

Host: yeah, yeah, there's some key information you want to collect about people, right, and

Data Collection and Product Information Management

Host: obviously their first name, last name, email address, phone number, those are great ones to collect, I mean pretty much standard, I know that from the check-in they're — you're automatically going to collect their driver's license information, right, so their address, date of birth, driver's license number, everything like that, and then when I come into the dispensary I check in, I buy — you know for me I love my pre-roll, so I'll get my pre-rolls, I can smoke those on the run, some sleeping edibles, and then I've now recently loved my cannabis drink, so I come in the store, I buy all those, you know, next week I can expect a message from Happy Cabbage or the retailer that, hey, Ken is having a vendor day here on Friday, and they're giving out 50 percent discounts on their drinks, he should come because he bought drinks in the past.

Andrew Watson: it's like — yeah, there you go, yeah, a hundred percent, yeah. one thing I'll say there as well, one other really important thing about first party data capture, is that it's awesome if you have all the information about the consumer, but what's even cooler is if you have all the information about the consumer and then all the information about every product that they've ever bought as well. so making sure that you are using a POS, or that you are at least curating and controlling how you're cataloging product in your system — you know, you mentioned Ken, right, making sure that you're using the brand features in your POS, making sure that you're appropriately using SKUs, product names, that kind of stuff on your e-commerce menu, connecting that through to your POS correctly, that's gonna give you a lot of rich information as well, so that categorically, hey, here are the Ken lovers, here are the beverage lovers, here are the sleepy edibles lovers, versus needing to have all this rich consumer data but then not knowing anything about anything they've ever bought, right, and that's also super key and super important.

Host: yeah, it's you know, when I go into a store that I have already shopped at, I expect them to know what I want, like I'm going in, I don't want to go to the dab area, it's not my deal, I want to see the flower, I want to see some of the edibles, and that's it, and if more advanced retailers, they'll typically have a bud tender with an iPad, and they'll say, oh, you know, based on what you purchased in the past, is what I think you'd like, and then take you in that direction, right.

Andrew Watson: that's what you'd expect, I mean that's what you get from, I would say, more advanced experiences within retail stores.

Host: right, yeah, and that's — extra, yeah.

Andrew Watson: yeah, we, because we have that data on the SMS side, we can profile the consumer and then figure out, hey, this is exactly the message to send to this consumer based on what they bought in the past, we also have an in-store recommendation tool as well that we sell with it, so it comes along with it, and that exactly, like a retailer can use that, look up the consumer on the iPad, and based on the real-time inventory at that store and that person's previous purchases, it'll actually give you a recommendation on exactly what to sell that person, that is going to optimize the price point, and we've actually seen people who use this tool to provide that recommendation based on your previous purchases see a 13 percent uplifted ticket size, because people are buying and selling things at full price, because they're more attuned to their preferences versus driving people to whatever product is discounted at this point in time. so again, other ways you can use first party data to augment the in-store interactions, don't just use it to market, but then, like, the full funnel is all the way from discovery of customer to that person making a purchase, right, and so make sure you're mapping and using the data all the way through.

Host: yeah, exactly, and, you know, you're collecting all this first-party data, Andrew, how can cannabis

How Should Cannabis Retailers Sort Their First-Party Data?

Host: retailers sort these audience segments and monetize them?

Andrew Watson: yeah, so that's where you can spend a lot of time sifting through this information, you can spend a lot of time doing analysis paralysis, and at the end of the day come up with a set of segments that you may believe exist, but perhaps don't have that much actual power when you go out and put them out into the world, right. and so what I always encourage is thinking of things based on recency, frequency, latency, that kind of stuff — how often is someone buying, how long has it been since they last purchased, that's going to be one of your largest driving factors when it comes to figuring out how much you should expect an audience to convert if they receive a message, or whether or not they need to be receiving messages or content, right, and so that could be your people who have fallen off, your 90-day shoppers, your weekly shoppers, your monthly shoppers, your bi-weekly shoppers, that kind of stuff. then from there I also heavily encourage thinking of product-category-based segments as well — what we have seen is that if you message somebody about a product category that they have purchased in the past, you will have twice — 2x, not 20 percent, not 30 percent, 200 percent — increase in the efficacy of that campaign, right. now, people who only buy edibles do not want to see your dab deal, it's just things they do not care about, right, I mean that leads to unsubscribes, it leads to low-to-no conversion, and then also conversion rates by category are very different — vapes, for example, we see very low conversion rates overall when people receive vape-based deals, for people who have vape preferences, dabs — people who buy dabs, a lot of primary dab preferences, who receive dab-based content, have significantly higher conversion rates than any other segment that we go after. and so I think thinking of things — how often people are buying and what they're buying — is really important, particularly in cannabis, when you think

Factors to Consider in Cannabis Consumer Behavior and Engagement

Andrew Watson: of the range of potency and the range of modality of products, is very very high, so you don't want to just do one-size-fits-all, because you're leaving potentially thousands, if not tens of thousands, on the table. and then, finally, I do encourage thinking of things like how far are people on their journey with knowing your brand and knowing your experience, right, and so that could be, like, who are your first-timers who have just tried you out and potentially are falling off, so you need to encourage them to use a delivery channel, for example, because they're far away, right, who are your high-frequent customers who for some reason keep coming back, and understanding their mindset and trying to think of ways that you can communicate to them in ways that are effective and that matter to the way that they're purchasing — you know, "we miss you," "happy Valentine's Day," "it's been a while," like, your birthday's an easy one, right.

Host: like birthday, that's a great one, that's a great one.

Andrew Watson: demographics, right, so thinking of, you know, we do see a lot of difference when it comes to seniors versus your Gen Z, particularly if your dispensary's close to a college, I will say, and so understanding that mix and types of things can help out a lot.

Host: it's good to know, is there any — I mean you highlight some campaigns, is there any that are your favorite, that are like, you know, people must include on theirs, I know

Effective Cannabis Marketing Campaigns and Strategies

Host: it's maybe perhaps like an advocacy or a deal campaign, like, is there any that really stick out to you?

Andrew Watson: yeah, yeah, some of the most effective ones that I've seen — hype brands, so brands that have been coming up a lot in your store, re-engaging people about those brands can heavily influence bringing people more back into the store, particularly with those discounts — now I don't necessarily encourage over-discounting on promotional stuff, but the fact of the matter is, you know, Jeeter is a good example, when you message Jeeter customers about the fact that there's a Jeeter deal, their Jeeter sales can quintuple in a day easily, right, and understand that with every one Jeeter product that they're buying, particularly if you have something like a delivery minimum, or while they're already in the store, they're also going to be adding on other products as well, right, and so the retailer on net is going to be making more from that than the brand, right, that's low-hanging fruit — so your top brands, making sure you're continuously communicating about your top brands is low-hanging fruit. and not to disparage more of the niche brands, but unfortunately the data does show that if you are messaging about niche brands at the expense of messaging about top brands, you may be leaving money on the table, and that's just the reality of the situation, because people are gonna be like, I don't know what this product is, they may be unsubscribing, they may get messaging fatigue. so that's one thing, and then the other thing I would say is regular nurture campaigns for lost customers by category are very effective, set it and forget it — make sure your edibles customers who haven't been back in the store in months are constantly receiving some messaging reminding them about edible deals, make sure your flower customers who haven't been back in months are constantly receiving some messaging about flower, those are the consistent things that each individual campaign may not get you a lot of money, but over the course of a month you can be bringing in a lot of revenue through them.

Host: yeah, a hundred percent, I know one that we love to do, and it's not really re-engaging, but it's more like a store launch campaign, we'll send people to a landing page before the store is open, let's say a month before, get them to join a VIP program, maybe offer like a 20 percent discount on the opening, build a list of 500 to 1,000 people, and then, on the day before grand opening, send out that message, and, you know, you have a line around the corner, and the owners of the dispensary — if it's a chain — they're gonna be happy, and it's not even the revenue, but it's the perception that the neighborhood and everyone around sees that you're a high demand retail store, right, and that people are standing in line for you, it's the same thing when you're standing outside a nightclub or something, right, people get FOMO.

Andrew Watson: yeah, they want to go check it out.

Host: yeah, word of mouth is an incredibly effective way to drive distance.

Andrew Watson: incredibly.

Host: yeah, yeah, exactly, and then, you know, I didn't mention it before, is combining SMS, email, and programmatic advertising to do more of an omnichannel marketing approach, and also, you know, SEO, to capture the keyword and search behavior — those are great ways for us to leverage this, and, you know, given the integration that we have between our two companies, I think it's

Display Ad Case Study: Impressive Results from Targeting with First-Party Data

Host: probably time we should show a case study here from one of the campaigns that we launched together, right?

Andrew Watson: yeah.

Host: so let's just look at this — so these numbers are impressive, let's just say the least. so we were in a test campaign, using the first party data from Happy Cabbage, and then we activated that data through MediaJel to reach these cannabis consumers through mainstream publishers, right, like your ESPN, your, the Chive, TMZ, Sports Illustrated, GQ, Salon.com, like all the mainstream publishers that you would expect. so Mankind spent 91 dollars on this campaign, so that is paying for placement on these publications, and they were able to generate 15,000 dollars in revenue from that, so I don't even know what the return on that is, but it's, it's very large, very large. so it's really an effective way for you to get in front of people, to engage with them, you know, you can send a text message, you can send out an email, maybe they don't respond, maybe they're distracted, maybe they're too high that day, whatever it might be, if that's the case, then you still want them to see ABC Dispensary, when they're playing Words With Friends, when they're trying to swipe right on a dating app, trying to find their next match. so a lot of opportunity there to get more awareness and just stay top of mind, really — it's the marketing rule of seven, the old-school way to say people need to see your brand seven times before purchasing, I think it's probably higher than that now, given all the attention economy and all the distractions that are available to us through our mobile phones nowadays, so something to consider. and then, looking at this campaign, we can actually see the lift from all the products, so you can look at the individual products, and you can see — oh, okay, well, I see Berry Fruit, True Fruit, that one's getting a lift for their 10-packs, Yuzu Lemon, here's another top product that's been effective in our marketing campaigns here. so, getting — oh, here we go, I just sorted by quantity, so it looks like Zen GMO, for this particular campaign, sold 20 units, Sativa 10-pack, Peach OZ, they're all — these are units that you maybe would have moved if you weren't running a programmatic display advertising campaign — my favorite strain of all time, Durban Poison, right there, number eight, eight sales. so just keeping that in mind, and you can also leverage this data to run a co-branded, co-marketing campaign, so let's say

SMS Case Study: Retargeting Campaign Powered by Customer Data

Host: Jeeter wants to move units at ABC Dispensary, they could run ads with Jeeter, I'll actually show you one — who am I running with, Jeeter's Green — I don't know if this one's active now, but you could showcase a brand as the actual advertisement, and work out a deal, or maybe you get a price break on purchasing, or maybe they give you some money to invest in marketing, so we're all in this together, and, you know, brands want to see the products sell through in a retail store, and retail stores want to get that product off the shelf as much as possible, so keep that in mind.

Andrew Watson: yeah, and that's what was really cool about this campaign, and I always like to say that SMS is just a channel, right, there are many different channels, this is not the be-all and end-all of channels, and what we've seen through usage of SMS is that the determination of efficacy is whether or not you leverage first party data to match people to when it's proper for them to be receiving content, and proper for them to be converting, and what content that is. what's really cool about this campaign is what we did, is we used the same machine learning technique that we use to identify who are the proper people to text, and we basically took that, applied it to the entire data set of the client, and then handed those device IDs over to MediaJel, and they were able to run this really awesome campaign against that data set, and get really great first party data back on individual transactions, right, and so that's why we were able to take 90 dollars and turn it into 15,000, exactly. but even within that, we got data back based on this that can help us retarget even further, right, and help lift that even further, and that's the beautiful thing about data science and linking the channels together, is you can make it learn, and you can make it grow, and make it become more effective over time, and to that point, if you have brand partners also engaged on this, right, being able to have a machine select out the individuals who are most primed to purchase Jeeter, who are falling off of that, feeding that selection of consumers back into the machine over and over and over again, so that those people are constantly receiving advertisements that are branded, that are bringing them back to your store to purchase that targeted brand, and we know, because again this like giant feedback loop, how to make that more effective over time, and so that's when we can take 90 and start turning it into 20, 30, 50 thousand, right, now I don't want to over-promise on anything, but

Host: yeah, like, yeah, I mean it's there, and you know, just looking at, we shoot for a 0.2 click-through rate on traditional campaigns, so by leveraging first-party data, it looks like we're about six times, seven times more effective at getting people to click on an ad. all right, so you're not wasting a bunch of money on showing ads — wait, I guess not waste, but you're not spending a bunch of money on people who are not your existing consumer, or not your consumer, you are advertising to people that have already purchased with you, and

Maximize Your Ad Spend with First-Party Data

Host: the way that we connect what you're doing, Andrew, on your side, to what we do, is an email address, right, an email address, mobile advertising ID as well, those are the two ways that we can connect, that's essentially like a social security number for your cell phone, right, so we can target ads based on that, and you can see here is a sample of, "20 off your next cannabis delivery," so this was, you know, you're trying to re-engage them to purchase again, looks like they probably lapsed for a while here, and then you can see the coverage, right, so, you know, from my experience, the top categories that are effective from

Find Opportunities to Re-engage Cannabis Customers On Their Favorite Media

Host: programmatic advertising is games, right, so everyone has games on their phone, no one likes to pay for them, so the way that these games monetize is through ads, all right, so games are huge, Reddit is on here, local — CBS, CBS San Diego, this is for Mankind in San Diego — podcasts, streaming, right, weather app, OfferUp, you know, it's basically — there's social media here, a little, memes, meme sites do really well, social media sites do really well, dating sites — Grindr is one of the highest performing publishers we work with, that category does really well, any dating, and especially gay dating does really well, podcast players — so you can just get an overview of the coverage, and at MediaJel we have 75,000 publishers in this, so there's quite a lot of opportunities for you to get in front of cannabis consumers, and you know that they're a cannabis consumer because we got that data from Happy Cabbage, and then we make sure that we distribute that data to all these publishers, which is all the publishers that accept advertising, and then looking at how they actually engage with you — so you can see, I'm going to change this to a five-day attribution window, so you can see that it ranges — some people only need to see you four times before they'll come by again, this person's eight times, if I open up their attribution window to all time, looks like some of these customers have to see us 50 times to come by again, all right. so the number of times the impression — anytime that they've seen it, right, someone saw this ad on 5/15, looks like they saw it four times on 5/15 on their Android phone, on their local breaking news app, and then they purchased on 5/19 for 42.37, and we know this because of an IP address match, so they come by again, and then now we're gathering more first party data about them and enriching the data set that we already have. so just a lot of value here to really maximize your return, this is a 419-to-one return on your investment, so it's absurd, to say the least.

Andrew Watson: well, yeah, there's just a lot of opportunity there to generate revenue for your retailer, I would say it's — the results are so compelling that if you're spending money on any retention SMS, retention email, retention loyalty, it's a no-brainer to also be spending money on retention-based programmatic, it's just a no-brainer, right, like, you're jumping all this money into discounting to try to bring people back, you might as well spend it on what the — like, the low-hanging fruit is, right.

Host: exactly, exactly, I mean, companies like Mercedes, BMW, they spend like 50 percent of their marketing budget on existing customers, right, so they have a specific brand positioning, right, luxury vehicles, and people want to associate or relate to and live that lifestyle, so there are reasons that they advertise in the ways that they do, it's, people want to feel part of the community, right, you know, the Tesla people are notorious, like if someone has a Tesla, they're part of the mission, the vision, right, so let's keep that in mind — what are — get me, Happy Cabbage campaigns, Andrew?

Andrew Watson: I'm Happy Cabbage campaigns, so, did I — did I catch you off guard there? I got you off guard — can you — could you be more specific?

Host: yeah, yeah, we can skip that one, we can skip that one. what about — I guess let's take a little break here, because I want to make sure that we answer any questions from the audience — so Ravi, Donald, Colleen, Zach, I don't know, do you have any questions for Andrew that you'd like us to answer live here? let me actually check social media too, because we may have some questions that people are asking on social, but feel free to ask your questions in the Q&A, which is at the bottom of the Zoom interface. just make sure there's no other questions here on LinkedIn — and nothing yet. all right, well, we'll give you some time, if you have any questions post them in the Q&A. all right, so, you want to talk a little bit about maybe the KPIs, right, which KPIs should a cannabis company be

Essential KPIs for Cannabis Marketing Success

Host: tracking, in order to know when to send out a re-engagement campaign?

Andrew Watson: yeah, so I'll actually start a little bit backwards in that — what I think, in terms of KPIs, right, what I think can sometimes get lost from folks is, at the end of the day, you're spending money on marketing to make money, but at the end of the day you are spending money in order to drive revenue, particularly when it comes to the bottom of the funnel marketing, which, you know, SMS campaigns that are very bottom level, right, discounts that you're giving in store, very bottom of the funnel, the primary KPI which you should be obsessed with is ROAS at the end of the day, it's how much money in revenue did you get from a dollar you spent on this campaign, right, particularly given how much low-hanging fruit — you know, these are people who already know your brand, they've already had experience, oftentimes they've had dozens of experiences beforehand, right, it's just a matter of really understanding, am I spending ten dollars, am I getting back a hundred, a thousand, ten thousand, exactly what those numbers are, right. now from there, in order to understand how you can increase or decrease ROAS, I think really strong, important KPIs to understand are conversion rate, right, so what percent of people, of a campaign, came back into a store within 24 hours, within seven days — we also will show recapture rates, things like that, so of all the people you sent out the campaign to, how many of them were latent and came back in, how many of them hadn't been back in the last 90 days and came back in, and how many dollars were associated with that, right. now, in terms of KPIs to figure out when the correct time and moment to send a campaign is, you actually take your results and you optimize your results, and so, for example, we have a tool that lets you see, based on time of day and day of week, what the appropriate time is, right — so let's say you've sent out a few re-engagement campaigns over the course of a week, you look back at that the next week and you say, okay, Saturday afternoon was by far the most effective time to send that campaign, let's resend on Saturday afternoon, let's double down on Saturday afternoons, right. understanding timing of brand deals, understanding timing of holidays, right, you don't want to burn through your list a week before Labor Day, when Labor Day categorically you've seen a certain amount of lift, and you want to double down on that lift — looking at your last year's data for your days leading up to 4/20, to understand how much of your sales in April came from the days before 4/20, things like that, right, I think are really important to understand, so that you're not just pressing the button constantly to send a whole bunch of messages constantly, because what you're just going to end up doing is burning your list, you're going to get some results, don't get me wrong, but they could be 10 percent of the results that you could potentially be getting.

Host: essentially, yeah, makes sense, makes — that same, make sure that you are dedicating a lot of time to re-engaging with your customers and making sure that they're coming back and that you retain them. Andrew, how do you calculate your customer retention rate?

How Do You Calculate Your Customer Retention Rate?

Andrew Watson: yeah, so — great question — yeah, so we think of, when we calculate retention, we think of it as — and, like I said, we use this 90-day mark, we came up with that through a statistical analysis of just, like, frequency and latency, and it's like a generic thing, but the number on retention I'd like to give people is, of customers who are acquired over a time period, what percent of them are still shopping 90 days later, or the converse of them, what percent of them are 90-day lost after that point in time, right. so if you want to say, okay, what is my retention of my consumers that are acquired in April, right, based on how often they're purchasing, by July you actually have a very good sense of what percent of your April consumers are still remaining consumers at your store, right, and then, basically, that's a rolling average that you can track to understand if you're getting better at retaining, or worse at retaining, the customers, right. and then, obviously, churn is just the converse of that, right, it's basically you either could be measuring what percent you're losing, or you could be measuring what percent you're retaining, they're the same concepts, right, just — it's a very important number to know, because people oftentimes measure how many new customers they're getting, but they're not checking how many of those new customers are staying, which is — yeah, not that good of a practice, because, like I said, you're spending so much money acquiring the new ones, you really want to make sure that you're keeping them.

Host: yeah, exactly, and I know that you put together a data brief, the campaign that we ran together, versus some of the other channels, do you want to share your screen and kind of

Case Study: Marketing Campaigns In-Store vs. Website vs. Weedmaps

Host: walk through that, about in-store versus website versus Weedmaps?

Andrew Watson: yeah, let me just open this up right now, and then let me just share my screen here, so you can see my screen, okay.

Host: I can — yeah, you can, yeah, I can see it.

Andrew Watson: yeah, so what we did is we basically looked at — so the point of that campaign with MediaJel was to go after previous customers who had fallen off, bring them back into the store. now what we looked at is how the customers who were reacquired through that campaign compared to customers who were reacquired from the website, in-store, or through Weedmaps, because they also have the integrated Weedmaps ordering, and what we found is that, when the customers are reacquired — we saw a pretty large uplift in average order value, basically, ticket size, a lot of people call it ticket size, right, where the campaign that we ran with MediaJel, people who are coming back into the store were spending about 122 on average, whereas people who are coming back — previous customers who then came back through Weedmaps, reordered through Weedmaps, where they were previously 90 days lost, spent about 95 on average, right. when we look at that a little bit deeper, right, and so we were just talking about retention, essentially what percent of those people, who it's been 90 days since they purchased, come back into the store, what percent of them are still purchasing within 90 days, right, because you want to think about the retention of the people you've recaptured, and what we saw is that the campaign that we ran, which was very targeted to individuals based on the first party data, 78 percent of those stayed purchasing at the store, right, versus 61 percent of those on Weedmaps. the other really important thing is that, by the way the campaign is designed, we're not double-dipping that much, right, the question is, are you basically inventing a new channel that is owned by, honestly, somebody else, that your consumers are constantly now reordering through — whereas with Weedmaps, what you see is that the customers who were recaptured through Weedmaps are reordering through Weedmaps, which means that you need to maintain a pay-to-play status in order to keep those customers who are already yours, whereas with this we were able to recapture them, bring more of them back into the store, and then we're not necessarily double-dipping, right, because they're coming back directly through in-store, through website, right, and then this is just — you can actually see how much of an uplift overall that is. and then, now, in terms of how long, and then how frequent it's been, what you can see is that, again, the online campaign we ran was very targeted, looking at people who are high-frequent shoppers who have dropped off on that 90-day window, and then bringing them back in, so we're not going hundreds of days without them being customers, right, we're able to essentially have the safety net, the safety net that as customers are falling out that bucket, or capturing them, putting them back in — which is why on average it was 93 days from basically seeing the ad to coming back in, their purchase, their most recent purchase, and then their repurchase, and then they were repurchasing after that every 12 days, whereas with Weedmaps you can see that it was 160 days, because it's not very targeted, right, it was a very long time, and when these people came back in, yes, they were more frequent, like they were purchasing basically once every week, but they were purchasing once every week through Weedmaps, basically Weedmaps owns the channel and you are kind of beholden to pay rent on that channel, right, whereas this is also going after people online, but it's incredibly more effective, and incredibly more tailored to the goals of the retailer, as opposed to saying, hey, rent a channel to maintain this way of customers ordering, and you have to basically pay dividends every month — instead, use your own data that you've already captured to bring them back into the store, and then maintain them through your own property, right, in which you're gonna get a higher average ticket size, you're gonna get more people coming back in faster, and then, overall, make more money.

Host: yeah, which is what really matters, right.

Andrew Watson: yeah.

Host: you want to own your customer data, you want to get them to purchase through your channels that you own, and you want to maximize the return, right, well, that's always the key takeaways that I got from that case study.

Andrew Watson: wonderful, I mean, thank you so much for taking time today, do you want to share with everyone where they can reach out to you and learn more about Happy Cabbage Analytics? actually, I'll pull up the website now and post it in the Zoom chat.

Closing Remarks

Andrew Watson: yeah, so while you do that, it's happycabbage.io, you can fill out a contact us form, there's also a chatbot, you can reach out directly, you can also email directly at sales@happycabbage.io if you're interested in getting in touch. we're really excited about rolling out with more partners, these online recapture campaigns with our MediaJel integration, 90 turns into over ten thousand dollars, really excited to bring those returns to more of the industry as well, and we're also, like I mentioned, rolling out soon more things on the inventory side, more things to round out activation data.

Host: okay, so, yeah, wonderful, well, thank you again for joining us today, we really enjoy working with you, obviously you're a great partner of ours, we appreciate the integration and the trust in us to really activate that data. just as a recap for everyone listening in here, this is the Cannabis Marketing Live podcast, we cover marketing trends and strategies for growing your cannabis business, that's programmatic advertising, SEO, paid search, SMS, email, push notifications, e-commerce, like pretty much anything and everything related to cannabis and tech. the next podcast will be airing live next Thursday, September 22nd, at 11 a.m. Pacific, 2 p.m. Eastern, and we'll be talking about bud tenders and how to not let them tank your cannabis brand, and that's going to be with Luna Stower, she's the chief impact officer at Inspired Hardware out of Los Angeles, so be sure to catch us next week for that, and we'll see you then. have a wonderful rest of your week, and we'll get you on the flip side.

Andrew Watson: thank you so much, of course, thanks.

‍

{}Introduction

Host: ciao everyone, welcome to the Cannabis Marketing Live podcast, where we cover the latest and most effective marketing trends and strategies to grow your cannabis dispensary, delivery service, or brand. I'm your host, and today we're discussing four ways to re-engage high-value cannabis — we're joined today by Andrew Watson, he's the CEO and founder of Happy Cabbage Analytics, welcome to the show, Andrew.

Andrew Watson: thank you for having me.

Host: yeah, of course, of course, before we kick it off today I'd like to give a shout out to our sponsor, MediaJel — MediaJel is the leading marketing platform helping cannabis brands reach consumers through their compliant ad network, with real-time reporting and analytics dashboard, as well as conversion tracking. well, let's kick it off here today, Andrew, tell me a little bit about your background and Happy Cabbage.

Andrew Watson: yeah, so, Happy Cabbage, we make revenue easy for cannabis retailers nationwide, by using the power of data science to take their first party data and turn it into revenue through predictive SMS marketing, inventory, product recommendations, and other tools that really make data and revenue generation easy for cannabis retailers. myself, I have a background in data science, coming from basically tech in San Francisco, and really excited to be able to bring a lot of those technology trends that happen outside of cannabis to cannabis, to really help us all make more money.

Host: really, yeah, I mean retailers need it more than ever, especially now, right, especially those California retailers, so anything we can get, we're gonna take. and you mentioned — was it recommended, or kind of automated text messaging — can you be more specific and give us an overview?

Andrew Watson: yeah, yeah, so what we do is we sit on top of your point of sale system, and we measure how often individual consumers are purchasing and what they're buying, and then we also have a prediction on, hey, if this set of a thousand consumers were to receive a text message today, this percent of them will convert, and therefore you'll make this amount of money, and we have learning models that are basically running every single day and measuring this trend across hundreds of retailers. and so what we can do is we can make it as easy as a click of a button, where we say, if you text these 300 people you will make two thousand dollars, would you like to send a message? right, and so it's very automated segmentation, we do provide you the tools to make it yourself if you want to, but basically we're like, look, the computer can do a much better job at selecting who you should message than a human can, objectively, so we just bring that technology to make it very powerful but very easy, which is really key.

Host: yeah, I remember doing, managing SMS campaigns like back in 2015-16, and I had to spend a ton of time segmenting customers, figuring out who bought what, figuring out what messages to send, so you pretty much cut out all that manual work, and it's all replaced with software and probably AI, right, and then you can do this like forecast modeling and everything?

Andrew Watson: that's nice, exactly.

Host: yeah, do the customers know about this, do they just log in and do this, or do they get an email like, hey, you have this opportunity —

Andrew Watson: yeah, yeah, so we have an interface, it's really simple, you just log in, and the first thing it says is, for example, one — you want to just show us?

Host: oh yeah, it's probably easier — yeah, I could just show you, if you would like to see.

Andrew Watson: so let me just really quickly pull this up.

Host: give us the tour, give us a tour.

Andrew Watson: yeah, yeah, yeah, once again — yeah, I didn't know that you did the forecasting and — yeah, so this is just a simple demo version of it, but what you can see here is what it's doing is the computer is — and it's basing it off of how well the computer did this previously, right, because we have the data on all the previous transactions and how often people buy, and then we also have the data on all the previous messages that have been sent millions of times, we know like what's the intersection, what's the overlap, what's the proper time for someone who's buying every seven days, or somebody's buying every 30 days, for them to receive a message. and then what we also have found is that messaging about what they actually prefer can double the efficacy of the message, and so what we've done here is we basically said, hey, we've automatically identified 340 people, we know they prefer flower, and we have a prediction, and all of this is dynamic to five percent of order again, you'll get seventeen hundred dollars in sales over seven days, you can write the message and you can click send, and then be done, basically. and so a lot of our customers who get the most effective results, the only thing that they do is they just blow through the opportunities and then they go about their day. we do also have — and this demo doesn't have it because we haven't loaded in the example data — but you can create your own custom campaigns, and you can get very specific into demographics and into product preferences and stuff, and what it'll do is it'll detect how well that campaign you came up with did, and then give you a prediction of how well it'll do again. so if you are wanting to do a bunch of your own analysis and build your own campaigns, you know, like, three or more times, eight or more times, you know, at like, buy in the afternoon, and you think that that's very powerful, the computer can say, okay, send it, and then based on how well that did, this is how well it'll do if you did it again.

Host: yeah, that's awesome, yeah, I know we have some customers that, you know, Mondays through Wednesdays are slow days, they want to try to get people in during those days, or through happy hours or something like that, just to even up the traffic flow in the store, because, you know, Thursday through Sunday it's just mayhem in some cases.

Andrew Watson: no, I like that.

Host: yeah, yeah, why should cannabis companies put effort in re-engaging

Why Should You Re-engage Lapsed Customers?

Host: latent or lapsed customers?

Andrew Watson: yeah, so the average retailer we run into, who has about two years of data, 50 — so half of all consumers who are opted in to receiving messaging haven't shopped in the previous 90 days. the average order frequency we see across all of the shops we deal with is approximately 25 days, so most consumers shop every 25 days. when half of your consumers who previously were shopping every 25 days now haven't shopped in the previous 90 days, that can be thousands, if not tens of thousands, of potential consumers who — you know their name, you know their address, you know what they bought, so you know what they like to buy, you know how often they buy, you know how long it's been since they bought, you know what time of day, you know day of week, you know everything about these consumers. and so by always having a strategy — and that's why one of our primary opportunities is "recapture lost customers" — a strategy of going after and bringing in those latent customers back in, basically, it's your most low-hanging fruit.

Host: yeah, right.

Andrew Watson: your cost of reacquiring those customers is minuscule compared to the cost of having to go out and acquire brand new customers off of, you know, services like Weedmaps or billboards or something like that, it's just, we're part of stable states, right.

Host: yeah, exactly, like it's kind of like, you know, you should be doing that, right, like in classic marketing theory, like you definitely should be constantly nurturing those who are falling off.

Andrew Watson: yeah, you should nurture them.

Host: you know, it's much more cost effective to engage with your existing customers, you have all these data points on them, activate that data, figure out a way to get them back in the store, whether it's working with your vendors to create a vendor day and then send people to — everyone that's interested in that brand for that vendor day, you can segment by that, right, get those people in the door, it's just so much money left on the table.

Andrew Watson: yeah, I mean, it's just business one-on-one, like, you have to nurture your existing customers, and within cannabis it's even more important, right, like email and SMS are fantastic, and push notifications as well, and you really just have to stay top of mind, so make sure that you're engaging with them and you're not spamming them, you're not sending the message every day, but make sure it's a curated message that's based on what they have purchased in the past, which is already built into Happy Cabbage, so all that stuff you can cut out, you cut out all the manual work you have to do, choose the opportunities, choose the vendors, brands that you want to highlight, make it work for yourself, and make sure that you re-engage with that 50 percent of people in that 90-day lapse time period, like that's just money on the table, and if you're not re-engaging with your

Re-Engaging Your Customers to Grow Revenue

Andrew Watson: customers, I don't know how you're going to grow your revenue month over month, like you're going to acquire new customers to grow your revenue, but you need to retain your existing customers, you get that compounded growth, so if you don't have that in place, you're just leaving money on the table.

Host: and, you know, I'll just add on, another couple strategies for re-engaging lapsed customers — now obviously at MediaJel we do programmatic advertising, so someone comes to the website, we pixel them, we continue to follow them around the web, anywhere they are, you know, if you think about going to Amazon, you're looking for shoes, you go to Instagram, you start seeing those shoes everywhere, it's the same thing, right, you want to follow your customers around, you can do that on programmatic, you can do that on Google as well. so just keep that in mind when you are figuring out ways to engage with your customers, and, you know, the best way is to engage with them in person, right, so they're in your store, make sure that you are prioritizing your top 20 percent of customers, that's another segment that I —

Andrew Watson: yeah, I feel like cannabis companies aren't leveraging that, like that top 10, 20 percent of your customers, you know, they probably don't take 25 days to shop, right, they're probably coming by every week, so have some type of messaging to make sure that they're coming back in the store, and then offer them loyalty points or some type of benefit to get them coming back, and whether it's vendor days, or what I always like to suggest is have an exclusive strain drop, or product drop, that's only available to the VIPs.

Host: yeah, it's just like — boom — it's a no-brainer, so you have all this first party data that you can leverage, so use it. Andrew, can you define what first party data is for us?

Defining First-Party Data in Cannabis

Andrew Watson: yeah, so first party data is data that, through the nature of your business, you've collected information about your consumers, and this could be if they're visiting your website, you know their IP address, you can know information about if they've logged into your menu, like what their email address is, information about that — if you are leveraging like an online audience or pixel type of stuff, you can normally map that to even more information. and then what's key in cannabis, which is one of the reasons why I started working in this industry as a data scientist, is your point of sale system also has all of this information about every single customer who's come through the door, right, so not only can you leverage all of the modern web technology that exists to be able to understand who's visiting the website, but you also have something that retailers outside of cannabis just don't have, which is every time someone buys something they need to check in, and with that you can connect the entire funnel all the way through, and you have very powerful information on what people want to buy. right, if you think about when you go into a store outside cannabis, every single time you buy something they ask for your email address, then you only give that 20 percent of the time maybe, whereas in cannabis, a hundred percent of the time you give your ID.

Host: so, oh yeah, with that kind of information, right, you have so much more power.

Andrew Watson: and that's why I think first party data in cannabis is about ten times, if not more, more valuable than it is outside of cannabis, because of just how rich and how complete it is.

Host: I agree, yeah, and, you know, there's a reason you go to Walgreens or CVS or Safeway or any store, there's a reason that they want you to join their loyalty program, there's a reason they're giving you discounts, because that data is worth more than oil, right.

Andrew Watson: yes, I don't know about nowadays because oil's gone through the roof, but, you know, I know that that's how Facebook, Amazon, Apple, and all these other companies, Tesla — you know, they're all data companies, right.

Host: yeah, they're — what's the difference between first party and second party data?

Andrew Watson: yeah, so second party data is basically data that you're buying from somebody else, very much, right, like I would say there's another company, right, that — this happens, New Frontier Data or something —

Host: yeah, exactly.

Andrew Watson: like they've mined data, honestly, they've mined other people's first party data, I'll put that out there, and they basically sold it back to you at a massive markup, because probably the people they got it from didn't realize how valuable it was in the first place, that's essentially second party data, and also, same thing, because there's so much rich first party data in cannabis, there happens to be a lot of second party resellers in cannabis as well.

Host: yeah, there's so much data available out there, you know, whether you're — number one, you should be capturing yourself. so we should probably talk a little bit about that right now, like, Andrew, what are some of the best ways you've seen first party data captured by a brand or a retailer?

Andrew Watson: yeah, so for retailers it's actually very simple, I will say, if you've made a good decision on a POS purchase, and those POS systems have very good, complete data — so for example a scanning technology on the ID that'll capture all that information off the ID, and then you have stops in place where you are collecting like phone number, email, other information — oftentimes POSes can even enforce your bud tenders or your receptionist to have to capture that information, which helps a lot, right, I would say that's the first line of attack, so to speak, on data collection, and that's where your largest source of truth data is coming from. on top of that, then your e-commerce system that you have as well, being able to make sure that you're collecting just as much data on that side as you're collecting with the in-store interactions, right, and so that means things like making sure you have an e-com checkout flow, an e-com system that is encouraging logins, that's encouraging people to fill out profiles, but isn't necessarily too onerous, so that you're not blocking potential transactions, right. and then from there, it's supplementary systems that can help add more contextual data on top of that, right, and so that's things like sending out surveys, getting feedback data, e-commerce systems that have reviews and stuff like that, you can actually connect a review to a person, you can get sentiment from that person, you can also do things like VIP programs, loyalty programs, that kind of stuff, can collect some more of that supplementary data. but I really do think that — and I'm saying this as someone who spent a lot of time doing healthcare data at work — the holy grail of data is having a robust connection from e-commerce to POS, where you're collecting all the information all along the way.

Host: yeah, yeah, there's some key information you want to collect about people, right, and

Data Collection and Product Information Management

Host: obviously their first name, last name, email address, phone number, those are great ones to collect, I mean pretty much standard, I know that from the check-in they're — you're automatically going to collect their driver's license information, right, so their address, date of birth, driver's license number, everything like that, and then when I come into the dispensary I check in, I buy — you know for me I love my pre-roll, so I'll get my pre-rolls, I can smoke those on the run, some sleeping edibles, and then I've now recently loved my cannabis drink, so I come in the store, I buy all those, you know, next week I can expect a message from Happy Cabbage or the retailer that, hey, Ken is having a vendor day here on Friday, and they're giving out 50 percent discounts on their drinks, he should come because he bought drinks in the past.

Andrew Watson: it's like — yeah, there you go, yeah, a hundred percent, yeah. one thing I'll say there as well, one other really important thing about first party data capture, is that it's awesome if you have all the information about the consumer, but what's even cooler is if you have all the information about the consumer and then all the information about every product that they've ever bought as well. so making sure that you are using a POS, or that you are at least curating and controlling how you're cataloging product in your system — you know, you mentioned Ken, right, making sure that you're using the brand features in your POS, making sure that you're appropriately using SKUs, product names, that kind of stuff on your e-commerce menu, connecting that through to your POS correctly, that's gonna give you a lot of rich information as well, so that categorically, hey, here are the Ken lovers, here are the beverage lovers, here are the sleepy edibles lovers, versus needing to have all this rich consumer data but then not knowing anything about anything they've ever bought, right, and that's also super key and super important.

Host: yeah, it's you know, when I go into a store that I have already shopped at, I expect them to know what I want, like I'm going in, I don't want to go to the dab area, it's not my deal, I want to see the flower, I want to see some of the edibles, and that's it, and if more advanced retailers, they'll typically have a bud tender with an iPad, and they'll say, oh, you know, based on what you purchased in the past, is what I think you'd like, and then take you in that direction, right.

Andrew Watson: that's what you'd expect, I mean that's what you get from, I would say, more advanced experiences within retail stores.

Host: right, yeah, and that's — extra, yeah.

Andrew Watson: yeah, we, because we have that data on the SMS side, we can profile the consumer and then figure out, hey, this is exactly the message to send to this consumer based on what they bought in the past, we also have an in-store recommendation tool as well that we sell with it, so it comes along with it, and that exactly, like a retailer can use that, look up the consumer on the iPad, and based on the real-time inventory at that store and that person's previous purchases, it'll actually give you a recommendation on exactly what to sell that person, that is going to optimize the price point, and we've actually seen people who use this tool to provide that recommendation based on your previous purchases see a 13 percent uplifted ticket size, because people are buying and selling things at full price, because they're more attuned to their preferences versus driving people to whatever product is discounted at this point in time. so again, other ways you can use first party data to augment the in-store interactions, don't just use it to market, but then, like, the full funnel is all the way from discovery of customer to that person making a purchase, right, and so make sure you're mapping and using the data all the way through.

Host: yeah, exactly, and, you know, you're collecting all this first-party data, Andrew, how can cannabis

How Should Cannabis Retailers Sort Their First-Party Data?

Host: retailers sort these audience segments and monetize them?

Andrew Watson: yeah, so that's where you can spend a lot of time sifting through this information, you can spend a lot of time doing analysis paralysis, and at the end of the day come up with a set of segments that you may believe exist, but perhaps don't have that much actual power when you go out and put them out into the world, right. and so what I always encourage is thinking of things based on recency, frequency, latency, that kind of stuff — how often is someone buying, how long has it been since they last purchased, that's going to be one of your largest driving factors when it comes to figuring out how much you should expect an audience to convert if they receive a message, or whether or not they need to be receiving messages or content, right, and so that could be your people who have fallen off, your 90-day shoppers, your weekly shoppers, your monthly shoppers, your bi-weekly shoppers, that kind of stuff. then from there I also heavily encourage thinking of product-category-based segments as well — what we have seen is that if you message somebody about a product category that they have purchased in the past, you will have twice — 2x, not 20 percent, not 30 percent, 200 percent — increase in the efficacy of that campaign, right. now, people who only buy edibles do not want to see your dab deal, it's just things they do not care about, right, I mean that leads to unsubscribes, it leads to low-to-no conversion, and then also conversion rates by category are very different — vapes, for example, we see very low conversion rates overall when people receive vape-based deals, for people who have vape preferences, dabs — people who buy dabs, a lot of primary dab preferences, who receive dab-based content, have significantly higher conversion rates than any other segment that we go after. and so I think thinking of things — how often people are buying and what they're buying — is really important, particularly in cannabis, when you think

Factors to Consider in Cannabis Consumer Behavior and Engagement

Andrew Watson: of the range of potency and the range of modality of products, is very very high, so you don't want to just do one-size-fits-all, because you're leaving potentially thousands, if not tens of thousands, on the table. and then, finally, I do encourage thinking of things like how far are people on their journey with knowing your brand and knowing your experience, right, and so that could be, like, who are your first-timers who have just tried you out and potentially are falling off, so you need to encourage them to use a delivery channel, for example, because they're far away, right, who are your high-frequent customers who for some reason keep coming back, and understanding their mindset and trying to think of ways that you can communicate to them in ways that are effective and that matter to the way that they're purchasing — you know, "we miss you," "happy Valentine's Day," "it's been a while," like, your birthday's an easy one, right.

Host: like birthday, that's a great one, that's a great one.

Andrew Watson: demographics, right, so thinking of, you know, we do see a lot of difference when it comes to seniors versus your Gen Z, particularly if your dispensary's close to a college, I will say, and so understanding that mix and types of things can help out a lot.

Host: it's good to know, is there any — I mean you highlight some campaigns, is there any that are your favorite, that are like, you know, people must include on theirs, I know

Effective Cannabis Marketing Campaigns and Strategies

Host: it's maybe perhaps like an advocacy or a deal campaign, like, is there any that really stick out to you?

Andrew Watson: yeah, yeah, some of the most effective ones that I've seen — hype brands, so brands that have been coming up a lot in your store, re-engaging people about those brands can heavily influence bringing people more back into the store, particularly with those discounts — now I don't necessarily encourage over-discounting on promotional stuff, but the fact of the matter is, you know, Jeeter is a good example, when you message Jeeter customers about the fact that there's a Jeeter deal, their Jeeter sales can quintuple in a day easily, right, and understand that with every one Jeeter product that they're buying, particularly if you have something like a delivery minimum, or while they're already in the store, they're also going to be adding on other products as well, right, and so the retailer on net is going to be making more from that than the brand, right, that's low-hanging fruit — so your top brands, making sure you're continuously communicating about your top brands is low-hanging fruit. and not to disparage more of the niche brands, but unfortunately the data does show that if you are messaging about niche brands at the expense of messaging about top brands, you may be leaving money on the table, and that's just the reality of the situation, because people are gonna be like, I don't know what this product is, they may be unsubscribing, they may get messaging fatigue. so that's one thing, and then the other thing I would say is regular nurture campaigns for lost customers by category are very effective, set it and forget it — make sure your edibles customers who haven't been back in the store in months are constantly receiving some messaging reminding them about edible deals, make sure your flower customers who haven't been back in months are constantly receiving some messaging about flower, those are the consistent things that each individual campaign may not get you a lot of money, but over the course of a month you can be bringing in a lot of revenue through them.

Host: yeah, a hundred percent, I know one that we love to do, and it's not really re-engaging, but it's more like a store launch campaign, we'll send people to a landing page before the store is open, let's say a month before, get them to join a VIP program, maybe offer like a 20 percent discount on the opening, build a list of 500 to 1,000 people, and then, on the day before grand opening, send out that message, and, you know, you have a line around the corner, and the owners of the dispensary — if it's a chain — they're gonna be happy, and it's not even the revenue, but it's the perception that the neighborhood and everyone around sees that you're a high demand retail store, right, and that people are standing in line for you, it's the same thing when you're standing outside a nightclub or something, right, people get FOMO.

Andrew Watson: yeah, they want to go check it out.

Host: yeah, word of mouth is an incredibly effective way to drive distance.

Andrew Watson: incredibly.

Host: yeah, yeah, exactly, and then, you know, I didn't mention it before, is combining SMS, email, and programmatic advertising to do more of an omnichannel marketing approach, and also, you know, SEO, to capture the keyword and search behavior — those are great ways for us to leverage this, and, you know, given the integration that we have between our two companies, I think it's

Display Ad Case Study: Impressive Results from Targeting with First-Party Data

Host: probably time we should show a case study here from one of the campaigns that we launched together, right?

Andrew Watson: yeah.

Host: so let's just look at this — so these numbers are impressive, let's just say the least. so we were in a test campaign, using the first party data from Happy Cabbage, and then we activated that data through MediaJel to reach these cannabis consumers through mainstream publishers, right, like your ESPN, your, the Chive, TMZ, Sports Illustrated, GQ, Salon.com, like all the mainstream publishers that you would expect. so Mankind spent 91 dollars on this campaign, so that is paying for placement on these publications, and they were able to generate 15,000 dollars in revenue from that, so I don't even know what the return on that is, but it's, it's very large, very large. so it's really an effective way for you to get in front of people, to engage with them, you know, you can send a text message, you can send out an email, maybe they don't respond, maybe they're distracted, maybe they're too high that day, whatever it might be, if that's the case, then you still want them to see ABC Dispensary, when they're playing Words With Friends, when they're trying to swipe right on a dating app, trying to find their next match. so a lot of opportunity there to get more awareness and just stay top of mind, really — it's the marketing rule of seven, the old-school way to say people need to see your brand seven times before purchasing, I think it's probably higher than that now, given all the attention economy and all the distractions that are available to us through our mobile phones nowadays, so something to consider. and then, looking at this campaign, we can actually see the lift from all the products, so you can look at the individual products, and you can see — oh, okay, well, I see Berry Fruit, True Fruit, that one's getting a lift for their 10-packs, Yuzu Lemon, here's another top product that's been effective in our marketing campaigns here. so, getting — oh, here we go, I just sorted by quantity, so it looks like Zen GMO, for this particular campaign, sold 20 units, Sativa 10-pack, Peach OZ, they're all — these are units that you maybe would have moved if you weren't running a programmatic display advertising campaign — my favorite strain of all time, Durban Poison, right there, number eight, eight sales. so just keeping that in mind, and you can also leverage this data to run a co-branded, co-marketing campaign, so let's say

SMS Case Study: Retargeting Campaign Powered by Customer Data

Host: Jeeter wants to move units at ABC Dispensary, they could run ads with Jeeter, I'll actually show you one — who am I running with, Jeeter's Green — I don't know if this one's active now, but you could showcase a brand as the actual advertisement, and work out a deal, or maybe you get a price break on purchasing, or maybe they give you some money to invest in marketing, so we're all in this together, and, you know, brands want to see the products sell through in a retail store, and retail stores want to get that product off the shelf as much as possible, so keep that in mind.

Andrew Watson: yeah, and that's what was really cool about this campaign, and I always like to say that SMS is just a channel, right, there are many different channels, this is not the be-all and end-all of channels, and what we've seen through usage of SMS is that the determination of efficacy is whether or not you leverage first party data to match people to when it's proper for them to be receiving content, and proper for them to be converting, and what content that is. what's really cool about this campaign is what we did, is we used the same machine learning technique that we use to identify who are the proper people to text, and we basically took that, applied it to the entire data set of the client, and then handed those device IDs over to MediaJel, and they were able to run this really awesome campaign against that data set, and get really great first party data back on individual transactions, right, and so that's why we were able to take 90 dollars and turn it into 15,000, exactly. but even within that, we got data back based on this that can help us retarget even further, right, and help lift that even further, and that's the beautiful thing about data science and linking the channels together, is you can make it learn, and you can make it grow, and make it become more effective over time, and to that point, if you have brand partners also engaged on this, right, being able to have a machine select out the individuals who are most primed to purchase Jeeter, who are falling off of that, feeding that selection of consumers back into the machine over and over and over again, so that those people are constantly receiving advertisements that are branded, that are bringing them back to your store to purchase that targeted brand, and we know, because again this like giant feedback loop, how to make that more effective over time, and so that's when we can take 90 and start turning it into 20, 30, 50 thousand, right, now I don't want to over-promise on anything, but

Host: yeah, like, yeah, I mean it's there, and you know, just looking at, we shoot for a 0.2 click-through rate on traditional campaigns, so by leveraging first-party data, it looks like we're about six times, seven times more effective at getting people to click on an ad. all right, so you're not wasting a bunch of money on showing ads — wait, I guess not waste, but you're not spending a bunch of money on people who are not your existing consumer, or not your consumer, you are advertising to people that have already purchased with you, and

Maximize Your Ad Spend with First-Party Data

Host: the way that we connect what you're doing, Andrew, on your side, to what we do, is an email address, right, an email address, mobile advertising ID as well, those are the two ways that we can connect, that's essentially like a social security number for your cell phone, right, so we can target ads based on that, and you can see here is a sample of, "20 off your next cannabis delivery," so this was, you know, you're trying to re-engage them to purchase again, looks like they probably lapsed for a while here, and then you can see the coverage, right, so, you know, from my experience, the top categories that are effective from

Find Opportunities to Re-engage Cannabis Customers On Their Favorite Media

Host: programmatic advertising is games, right, so everyone has games on their phone, no one likes to pay for them, so the way that these games monetize is through ads, all right, so games are huge, Reddit is on here, local — CBS, CBS San Diego, this is for Mankind in San Diego — podcasts, streaming, right, weather app, OfferUp, you know, it's basically — there's social media here, a little, memes, meme sites do really well, social media sites do really well, dating sites — Grindr is one of the highest performing publishers we work with, that category does really well, any dating, and especially gay dating does really well, podcast players — so you can just get an overview of the coverage, and at MediaJel we have 75,000 publishers in this, so there's quite a lot of opportunities for you to get in front of cannabis consumers, and you know that they're a cannabis consumer because we got that data from Happy Cabbage, and then we make sure that we distribute that data to all these publishers, which is all the publishers that accept advertising, and then looking at how they actually engage with you — so you can see, I'm going to change this to a five-day attribution window, so you can see that it ranges — some people only need to see you four times before they'll come by again, this person's eight times, if I open up their attribution window to all time, looks like some of these customers have to see us 50 times to come by again, all right. so the number of times the impression — anytime that they've seen it, right, someone saw this ad on 5/15, looks like they saw it four times on 5/15 on their Android phone, on their local breaking news app, and then they purchased on 5/19 for 42.37, and we know this because of an IP address match, so they come by again, and then now we're gathering more first party data about them and enriching the data set that we already have. so just a lot of value here to really maximize your return, this is a 419-to-one return on your investment, so it's absurd, to say the least.

Andrew Watson: well, yeah, there's just a lot of opportunity there to generate revenue for your retailer, I would say it's — the results are so compelling that if you're spending money on any retention SMS, retention email, retention loyalty, it's a no-brainer to also be spending money on retention-based programmatic, it's just a no-brainer, right, like, you're jumping all this money into discounting to try to bring people back, you might as well spend it on what the — like, the low-hanging fruit is, right.

Host: exactly, exactly, I mean, companies like Mercedes, BMW, they spend like 50 percent of their marketing budget on existing customers, right, so they have a specific brand positioning, right, luxury vehicles, and people want to associate or relate to and live that lifestyle, so there are reasons that they advertise in the ways that they do, it's, people want to feel part of the community, right, you know, the Tesla people are notorious, like if someone has a Tesla, they're part of the mission, the vision, right, so let's keep that in mind — what are — get me, Happy Cabbage campaigns, Andrew?

Andrew Watson: I'm Happy Cabbage campaigns, so, did I — did I catch you off guard there? I got you off guard — can you — could you be more specific?

Host: yeah, yeah, we can skip that one, we can skip that one. what about — I guess let's take a little break here, because I want to make sure that we answer any questions from the audience — so Ravi, Donald, Colleen, Zach, I don't know, do you have any questions for Andrew that you'd like us to answer live here? let me actually check social media too, because we may have some questions that people are asking on social, but feel free to ask your questions in the Q&A, which is at the bottom of the Zoom interface. just make sure there's no other questions here on LinkedIn — and nothing yet. all right, well, we'll give you some time, if you have any questions post them in the Q&A. all right, so, you want to talk a little bit about maybe the KPIs, right, which KPIs should a cannabis company be

Essential KPIs for Cannabis Marketing Success

Host: tracking, in order to know when to send out a re-engagement campaign?

Andrew Watson: yeah, so I'll actually start a little bit backwards in that — what I think, in terms of KPIs, right, what I think can sometimes get lost from folks is, at the end of the day, you're spending money on marketing to make money, but at the end of the day you are spending money in order to drive revenue, particularly when it comes to the bottom of the funnel marketing, which, you know, SMS campaigns that are very bottom level, right, discounts that you're giving in store, very bottom of the funnel, the primary KPI which you should be obsessed with is ROAS at the end of the day, it's how much money in revenue did you get from a dollar you spent on this campaign, right, particularly given how much low-hanging fruit — you know, these are people who already know your brand, they've already had experience, oftentimes they've had dozens of experiences beforehand, right, it's just a matter of really understanding, am I spending ten dollars, am I getting back a hundred, a thousand, ten thousand, exactly what those numbers are, right. now from there, in order to understand how you can increase or decrease ROAS, I think really strong, important KPIs to understand are conversion rate, right, so what percent of people, of a campaign, came back into a store within 24 hours, within seven days — we also will show recapture rates, things like that, so of all the people you sent out the campaign to, how many of them were latent and came back in, how many of them hadn't been back in the last 90 days and came back in, and how many dollars were associated with that, right. now, in terms of KPIs to figure out when the correct time and moment to send a campaign is, you actually take your results and you optimize your results, and so, for example, we have a tool that lets you see, based on time of day and day of week, what the appropriate time is, right — so let's say you've sent out a few re-engagement campaigns over the course of a week, you look back at that the next week and you say, okay, Saturday afternoon was by far the most effective time to send that campaign, let's resend on Saturday afternoon, let's double down on Saturday afternoons, right. understanding timing of brand deals, understanding timing of holidays, right, you don't want to burn through your list a week before Labor Day, when Labor Day categorically you've seen a certain amount of lift, and you want to double down on that lift — looking at your last year's data for your days leading up to 4/20, to understand how much of your sales in April came from the days before 4/20, things like that, right, I think are really important to understand, so that you're not just pressing the button constantly to send a whole bunch of messages constantly, because what you're just going to end up doing is burning your list, you're going to get some results, don't get me wrong, but they could be 10 percent of the results that you could potentially be getting.

Host: essentially, yeah, makes sense, makes — that same, make sure that you are dedicating a lot of time to re-engaging with your customers and making sure that they're coming back and that you retain them. Andrew, how do you calculate your customer retention rate?

How Do You Calculate Your Customer Retention Rate?

Andrew Watson: yeah, so — great question — yeah, so we think of, when we calculate retention, we think of it as — and, like I said, we use this 90-day mark, we came up with that through a statistical analysis of just, like, frequency and latency, and it's like a generic thing, but the number on retention I'd like to give people is, of customers who are acquired over a time period, what percent of them are still shopping 90 days later, or the converse of them, what percent of them are 90-day lost after that point in time, right. so if you want to say, okay, what is my retention of my consumers that are acquired in April, right, based on how often they're purchasing, by July you actually have a very good sense of what percent of your April consumers are still remaining consumers at your store, right, and then, basically, that's a rolling average that you can track to understand if you're getting better at retaining, or worse at retaining, the customers, right. and then, obviously, churn is just the converse of that, right, it's basically you either could be measuring what percent you're losing, or you could be measuring what percent you're retaining, they're the same concepts, right, just — it's a very important number to know, because people oftentimes measure how many new customers they're getting, but they're not checking how many of those new customers are staying, which is — yeah, not that good of a practice, because, like I said, you're spending so much money acquiring the new ones, you really want to make sure that you're keeping them.

Host: yeah, exactly, and I know that you put together a data brief, the campaign that we ran together, versus some of the other channels, do you want to share your screen and kind of

Case Study: Marketing Campaigns In-Store vs. Website vs. Weedmaps

Host: walk through that, about in-store versus website versus Weedmaps?

Andrew Watson: yeah, let me just open this up right now, and then let me just share my screen here, so you can see my screen, okay.

Host: I can — yeah, you can, yeah, I can see it.

Andrew Watson: yeah, so what we did is we basically looked at — so the point of that campaign with MediaJel was to go after previous customers who had fallen off, bring them back into the store. now what we looked at is how the customers who were reacquired through that campaign compared to customers who were reacquired from the website, in-store, or through Weedmaps, because they also have the integrated Weedmaps ordering, and what we found is that, when the customers are reacquired — we saw a pretty large uplift in average order value, basically, ticket size, a lot of people call it ticket size, right, where the campaign that we ran with MediaJel, people who are coming back into the store were spending about 122 on average, whereas people who are coming back — previous customers who then came back through Weedmaps, reordered through Weedmaps, where they were previously 90 days lost, spent about 95 on average, right. when we look at that a little bit deeper, right, and so we were just talking about retention, essentially what percent of those people, who it's been 90 days since they purchased, come back into the store, what percent of them are still purchasing within 90 days, right, because you want to think about the retention of the people you've recaptured, and what we saw is that the campaign that we ran, which was very targeted to individuals based on the first party data, 78 percent of those stayed purchasing at the store, right, versus 61 percent of those on Weedmaps. the other really important thing is that, by the way the campaign is designed, we're not double-dipping that much, right, the question is, are you basically inventing a new channel that is owned by, honestly, somebody else, that your consumers are constantly now reordering through — whereas with Weedmaps, what you see is that the customers who were recaptured through Weedmaps are reordering through Weedmaps, which means that you need to maintain a pay-to-play status in order to keep those customers who are already yours, whereas with this we were able to recapture them, bring more of them back into the store, and then we're not necessarily double-dipping, right, because they're coming back directly through in-store, through website, right, and then this is just — you can actually see how much of an uplift overall that is. and then, now, in terms of how long, and then how frequent it's been, what you can see is that, again, the online campaign we ran was very targeted, looking at people who are high-frequent shoppers who have dropped off on that 90-day window, and then bringing them back in, so we're not going hundreds of days without them being customers, right, we're able to essentially have the safety net, the safety net that as customers are falling out that bucket, or capturing them, putting them back in — which is why on average it was 93 days from basically seeing the ad to coming back in, their purchase, their most recent purchase, and then their repurchase, and then they were repurchasing after that every 12 days, whereas with Weedmaps you can see that it was 160 days, because it's not very targeted, right, it was a very long time, and when these people came back in, yes, they were more frequent, like they were purchasing basically once every week, but they were purchasing once every week through Weedmaps, basically Weedmaps owns the channel and you are kind of beholden to pay rent on that channel, right, whereas this is also going after people online, but it's incredibly more effective, and incredibly more tailored to the goals of the retailer, as opposed to saying, hey, rent a channel to maintain this way of customers ordering, and you have to basically pay dividends every month — instead, use your own data that you've already captured to bring them back into the store, and then maintain them through your own property, right, in which you're gonna get a higher average ticket size, you're gonna get more people coming back in faster, and then, overall, make more money.

Host: yeah, which is what really matters, right.

Andrew Watson: yeah.

Host: you want to own your customer data, you want to get them to purchase through your channels that you own, and you want to maximize the return, right, well, that's always the key takeaways that I got from that case study.

Andrew Watson: wonderful, I mean, thank you so much for taking time today, do you want to share with everyone where they can reach out to you and learn more about Happy Cabbage Analytics? actually, I'll pull up the website now and post it in the Zoom chat.

Closing Remarks

Andrew Watson: yeah, so while you do that, it's happycabbage.io, you can fill out a contact us form, there's also a chatbot, you can reach out directly, you can also email directly at sales@happycabbage.io if you're interested in getting in touch. we're really excited about rolling out with more partners, these online recapture campaigns with our MediaJel integration, 90 turns into over ten thousand dollars, really excited to bring those returns to more of the industry as well, and we're also, like I mentioned, rolling out soon more things on the inventory side, more things to round out activation data.

Host: okay, so, yeah, wonderful, well, thank you again for joining us today, we really enjoy working with you, obviously you're a great partner of ours, we appreciate the integration and the trust in us to really activate that data. just as a recap for everyone listening in here, this is the Cannabis Marketing Live podcast, we cover marketing trends and strategies for growing your cannabis business, that's programmatic advertising, SEO, paid search, SMS, email, push notifications, e-commerce, like pretty much anything and everything related to cannabis and tech. the next podcast will be airing live next Thursday, September 22nd, at 11 a.m. Pacific, 2 p.m. Eastern, and we'll be talking about bud tenders and how to not let them tank your cannabis brand, and that's going to be with Luna Stower, she's the chief impact officer at Inspired Hardware out of Los Angeles, so be sure to catch us next week for that, and we'll see you then. have a wonderful rest of your week, and we'll get you on the flip side.

Andrew Watson: thank you so much, of course, thanks.

‍