Product Release: How to Maximize Sales with Multi Touch Attribution Reports

[ {Introduction}

okay hello happy thursday everyone welcome to the MediaJel podcast where we cover the latest marketing trends and strategies that are most effective in growing your cannabis dispensary delivery service or brand MediaJel has a compliant ad network of 75 000 pub plus publishers including mainstream news sites we love the meme sites dating gaming music streaming apps and they all welcome cannabis ads we help cannabis companies here at MediaJel through paid search seo and programmatic display advertising to drive awareness increase foot traffic and e-commerce sales i am your host guillermo bravo today we'll be highlighting our product release and how to use attribution reporting to maximize revenue thank you for joining us today all right so what we're going to be spending a lot of time today doing is um you know i'm going to be sharing my screen i'm going to be going through uh a bunch of different campaigns here uh so give me a second here and let me share my screen all right all right so i've got my share screened over here so you know today we're going to spend a lot of time talking about attribution we really we really want to understand where revenue comes from you know this is pivotal for allocating your marketing budget yet cannabis e-commerce platforms don't always make this clear and some don't even give the option to review your data we're committed here at MediaJel to providing our clients with data and insights they require to make smarter marketing decisions and attribute revenue to each marketing campaign that we run so you know we really want to make sure that we're actively tracking uh where your marketing spend is going and what that leads to uh when you deter when you can determine your sales and where they're coming from and how well your digital marketing campaigns are fair and you can make smarter marketing decisions and invest where it matters let's look at brand dispensary and delivery service campaigns and dive in deep to the reports through the MediaJel dashboard so the first one i'm going to look at here and please please feel free to ask your questions by clicking in the chat box at the bottom of the zoom interface or if you click the q a you can ask questions there so be feel free to ask away and i'll be sure to ask answer your questions live during the session

{Roll Up View}

all right so this is the what we call the roll up view so this is a view of your entire performance uh across all three marketing channels so let me just dive in deep here all right so first thing we're looking at here this is the home home base of your dashboard so this is looking at all campaigns running across display paid search and seo so what this is looking at is your overall impressions served across all channels the the budget that was spent uh in that timeline so this is all time and the revenue that was generated from that campaign and then the return on ad spend so for each of the reports that we're going to be covering today we really want to you know confirm uh what the overall objective of the campaign was you know how do we measure success and then how we optimize the campaigns so there's a lot of different things that we're going to cover today um so really uh you know get comfortable and you know ask questions away all right so let's kind of dive down a little bit so for this campaign this is a cannabis retail chain and their goal was to uh increase e-commerce sales for their retail stores drive walk-ins and increase increase just revenue for the business so the first thing we're looking at here is the overall summary so impressions

{Overall Summary View}

served budget revenue return on ad spend so this looks like it was close to a 20 to one on the return on ad spend if we look at the display area all time we spent eighty four thousand dollars on display campaigns for this for this retailer from that we generated 4.7 million dollars and attributed multi-touch attribution revenue and that's through 52 000 plus transactions so that was at a click through impressions total impression served for the campaign was just over 12 million impressions that generated 36 609 clicks which is a 0.304 click-through rate so that's the number of people who click on the campaign as a percentage of the total impressions that's the click-through rate uh so this is a very healthy campaign here uh looks like it's uh 56-1 and the way that you would attribute your return on ad spend would be the total revenue generated from the campaign which is 4.7 million and then you just divide that by the 84 000 and that's where you'll get the the 56 56

{Paid Search Section}

next we're going to go through the paid search so paid search that's going to be leveraging google ads or bing ads to pay for placement on search engines and you know with the goal of driving last click revenue and that's a little different than display so i'll cover that a little bit here shortly so for the paid search campaign we served over a million impressions we generated 84 285 clicks that was a click-through rate of almost eight percent it was a cost per click so the number the amount it cost every time that someone clicks on those ads was a dollar and 22 cents and then we generated 516 000 in revenue from that campaign so that's another look at uh campaign here and i can actually just generate i can see how much we invested on this so 1.22 times 84 285 clicks so we spent just over a hundred thousand dollars on this campaign and we generated 516 000 in last click revenue so that is in just over a five to one return on investment or return on ad spend for this channel next we're looking at seo

{SEO Section}

uh so for this campaign we generated 3.2 million total sessions with 1.4 million new users we generate 000 323.000 with the e-commerce conversion rate of close to 20 percent which is really really high in uh you know this campaign they have a very solid uh brand equity and with that's been attributed to a high e-commerce conversion rate and they have really spent a lot of time curating their product catalog and working their pricing model to to be right and fit with the market which yields this high conversion rate and then from the sdo all time they generate close to 32 million dollars in sales so you can see the the sheer volume across three channels so 32 million dollars in revenue from seo all time 516 000 in revenue from paid search and then 4.7 million in influence sales for display programmatic advertising all right so that's just a look at the roll up view so you know if you you're working with MediaJel and you you want to just get an overview of kind of where you stand today boom you can have that you can update this here you can just change the date range and then this will update all the metrics below so that let's just do i just leave it there so that's kind of the roll-up view all right let's look at a dispensary e-commerce campaign so you know first let's define you know what the overall objective is for this

{Campaign Example: Dispensary eCommerce Campaign}

campaign so you know this partner came to us and they were investing nine thousand dollars over 90 days uh in display advertising so this is the display campaign you know the goal of the campaign was to eq increase e-commerce sales for their retail stores and this was to the dutchy menu so this retailer he was using the dutchy menu and they really wanted to drive online pre-orders uh and also drive walk-ins for their store uh so that was the focus of this campaign so with a 9 000 investment this partner was able to generate 30 322 000 in influence revenue and i'll kind of cover what that means here so when we're looking at when we're talking about display advertising and we're we're thinking about multi-touch attribution and in when we say influence sales what we're saying is that we showed a banner ad to this potential consumer and later down the line they made a purchase through our e-commerce menu uh it could be like five days ten days later and i'll show you kind of a a flow of that transaction here but that's multi-touch attribution and that's click through uh transactions as well as another way you can explain it so this is the summary page for this partner uh they served not 1.3 million impressions for that 9 000 investment and i believe that that was at a cpm rate of seven dollars but let me just calculate that here for you yeah this is a seven dollar cpm so we served impressions to 1.3 million uh we served impressions to 1.3 million i guess instances uh to i'm not sure how many actual consumers but they saw our ads 1.3 million times from this they clicked five thousand times they clicked in five thousand times of the session and of this we generate 3988 transactions with a multi-touch attribution revenue or influence sales of 322 000. and from here we can see the clicks in blue and the click-through rate ratio in teal and all of this is customizable so i'll just kind of change this by transaction total so the life of the campaign ran 90 days but we do track sales 30 days after the campaign stopped serving ads because that is our attribution window and the attribution window is the time between the last the last time that the consumer saw one of your ads to the day they purchased and we'll dive in deep a little bit more into that when we go into the transactions tab here you can look at transaction total we can look at a total impression served we can look at clicks and we can compare that to transaction or transaction count we can change the time frame to be you know last 30 days this quarter last quarter so there's a lot of ability to change the the date range and compare a metric side by side if we go down a little bit more you can see the top 10 publishers here so for this campaign we have looks like some games some public some publishers he has some dating sites some social media sites some meme sites uh senior sites as well and then we have the top 10 creatives on the right here so this is a look at the summary page if you're just looking at the overall metrics for this campaign and we're going to dive into the display side all right so for this for this uh dispensary ecommerce campaign we're going to have a look at the impressions clicks and click-through rate for the campaign uh so this campaign ran from 2 9 to 5 11. total impression serves just over 1.3 million with generating 5000 plus clicks with the click-through rate of 0.38 here we can sort this by impressions click-through rate or clicks and click here so you can see the impression served here in blue and then then total clicks generated in teal so you can see how this changes and it's just based on how people interact with the actual ad so if they click on it you're going to see a point here in the teal and you can see this kind of varies you can sort this by day of the week so it looks like saturday and thursday are the highest days for this partner and then device type so it looks like most of the clicks came from mobile devices which is pretty common for programmatic display campaigns and then tablets 345 and then also personal computers and then a combination of mobile and tablet so you can see the breakup of all the different ad placements that are available for display advertising we can also we also have one from connected tv it looks like and then set top boxes had zero clicks so we do have the ability to run these ads on connected tv in gaming boxes phones tablets and desktops here you can see a breakup of all the impressions and clicks per ad type and size so this is a 320 by 50 campaign we had 200 000 impressions so that's the number of times the ad was shown to a user and then of those impressions 70 751 people clicked on the ad with the click-through rate of 0.36 which is very very high we typically look for an average of 0.20 for the click-through rate and this is far performed for far outperforming uh the typical campaign so point two is a kind of an industry standard for click-through rate and this is two so that's the highest uh and you can see how this kind of changes down the line all right so this campaign is doing really well let's take a look at the publishers so as you can see there is a ton of publishers within within the campaign i want to actually sort by click-through rate all right this is a campaign running i believe all throughout throughout the country across five states so they really have a widespread for the distribution of the publishers so you can see the publishers by total volume of clicks uh temple cool was the number one ifunny which is a meme site and video site was number two got some gaming we got some song streaming we got some dating we got some humor over wolf sudoku oc register so you can see the the breakup of all the different publishers here and the one thing to keep in mind for display advertising is like to not get really stuck on the publishers uh the the one thing to remember is that we're targeting known cannabis consumers or we're geo fencing for this campaign we were geofencing all their competitors we call that prospecting and you know the goal of this was to take competitors consumers so anyone who walks into one of our competing retailers uh they'll get placed into our audience and we can do a look back for the last six months of the campaign to see how many people walked into our competing dispensaries that are nearby us and then we can add those to our audience and target those users if they walk into a rg event so let's just say we want to target people to walk in within a 10 mile radius of our retail store and anyone who has been to a cannabis dispensary over the last six months we want them to see our ads and encourage them to come into our store and then we'll just dive in deep on optimization of the campaigns so you know you may want to see your ads on only on espn so if you want to optimize your campaigns for conversion rates you know one the first thing we do is publisher placement so let's say you want to see your ads on espn rolling stones or [Music] condensed websites but that might not be where most your valuable audiences hang out on the web so you may want to shift the focus to publishers that you know will make an impact on click-through rates we're big on mobile-only ads seeing that you know a majority of cannabis consumers do their shopping through their mobile device so if you're targeting a specific device and your campaign isn't serving well you know we may want to consider cross-device targeting and we we enable this by default so yeah showing ads on your mobile device tablet desktop tv we want to make sure that we have full coverage across the board for that cannabis consumer and if we don't have the the scale and the ability to reach you know these the impressions that we're looking to achieve we can expand our audience targeting so understanding your audience their browsing and buying behaviors is key we're huge on hyper local marketing the practice of using location-based targeting to narrow in on consumers in specific locations you know to increase the odds of converting those users into consumers so you can shift the targeting area or layer in new audience data to expand your reach so let's say if we're doing a five mile radius from our store maybe we open that up to seven miles or ten miles to allow us to reach new consumers maybe we add new competitors to our targeting we can also buy third-party data from let's say you can layer on like a new frontier data or some of these other data partners and activate those data points through our through our campaigns or if you have like a crm that you're leveraging like a spring bag or a happy cabbage or alpine iq you know you can export the emails from your crm and we can turn those emails into maids mobile advertising ids and then from there we can activate those maids with targeted campaigns for display advertising so that's another way that we can do campaign optimizations and then sometimes it's you know if the click-through rates are low which is not the case in this campaign uh you maybe you you may need to make some updates to the content or call to actions in your creative so let's say let's look at one of these low so this these are all low below a 0.2 0 click-through rate so we may want to update the call to action maybe change the color shift the copy to increase click-through rate and engagement so just keep in mind that digital ads can appear almost anywhere in the online space so it could be media or e-commerce sites news feeds videos audio or ctv streaming so experimenting with static banners versus gist versus pop-ups or in-stream ads whether it's before during or after video will all give us valuable insights as to what connects with potential customers the best and then if you are running the ad on a on a paid search campaign we'll talk about that later but you you may want to shift your keyword strategy so while it's understandable that you're bidding on like some of the most popular engineering keywords like dispensing near me or or buy wheat online sometimes that could be a losing strategy as the return on ad spend is is not really up to par so that's something that you'll want to look at and see if there's any opportunities uh to become more competitive as the cannabis market becomes you know tighter and more and there's just more stores popping up we're seeing a trend uh towards greater uh specif specificity and by choosing you know let's say long tail keywords you could start to outperform uh better funded but less nimble keyword strategies that's just something to keep in mind all right now let's take a look at the [Music] attribution for the campaigns so first let's talk about the influence sales so you know multi-touch attribution uses device level data that shows impressions that have been served to a device that have also made a transaction to the you know through the uh e-commerce menu so that's ex that's exactly uh you know the definition of multi-touch attribution is like we served an ad to a user later down the line they made a purchase so let's go down a little bit here so for this campaign remember they spent just over nine they spent nine thousand dollars on this campaign uh so we were able to generate 322 000 influence sales off that 9 000 investment this was to 2193 unique customers that place 3988 unique transactions so some of these customers actually paste multiple transactions and this is a return on ad spend of 35 to 1. this is really really really high performing campaign for a display at programmatic advertising and when we're looking at the

{Sorting Transactions}

transactions here you know we can sort this by day of the week by whoops let me just refresh that here one moment just refreshing the page all right perfect so continuing down i can change this to day of the week so it looks like friday is our number one day for total transactions and revenue followed by saturday at 599 with 50 000 close to 50 000 in revenue and then thursday at 560 transactions with 43 000 in revenue so if we go down here let's go to uh this campaign so this is a great example so what this is saying is that this actual cannabis consumer was served an ad starting on 4 19 so just just before 4 20 they were served this ad on an iphone and these were the publishers that they were served on so they have an iphone they were served nine ads across all these different publishers so background checks uh crypto mining game uh theweek.com and then they went to the shopping cart dutchy this is the actual duchy order id the last impression that that was served to this consumer was on 5 8 at 1 58 pm local time they were served nine impressions they made a purchase through the dutchy menu on 5 12 looks like at 8 32 pm they spent 45 dollars and 60 cents and we knew that we know this through a user match so this is a combination of the user agent and ip address to produce the hash so let me talk a little bit about the different ways that we match a transaction to an impression so that you have a better idea of you know how that how we address um you know matching purchases and devices with ip addresses cookies or email addresses so the deterministic match also known as uh as explicit matching is a method to find the exact match between two sets of data to identify the same user across different channels and devices users are matched based on first party data details with the following identifiers so email address phone number login details whether it's like your username address date of birth etc so all this is gathered to to match a transaction to a user and then probabilistic match also known as implicit matching compares several data points together to identify the same user across multiple devices channels or platforms in general a knowledge database and predicting algorithms are used to data used by data aggregators and identify an identity solution providers during the identification process so in probable probabilistic matching devices are linked by looking into the following data points this is ip address wi-fi networks device fingerprinting screen resolution operating system device type and so on so by combining several data points a user is identified and the statistical likelihood helps to find the same user on different devices um so that's just a high level of this and this is identified here on the right side so these different icons state different match types so i'm just going to expand this a little bit all right so it looks like most of these are probabilistic match see if there's any others here we go ip address match that this user was served an add-on 225 225 on an iphone and they purchased some 411 for 28.80 and we know this because of ip address match and you can see that the date between the last impression and the order is almost a month and two weeks so you can actually change the attribution window window here uh let's just say to two days and you can see that the influenced sales and custom consumers and transactions updates at the top so we don't actually say uh which attribution window you should use uh we allow you to choose your attribution window so you can make the best decisions for your marketing team uh so you can see a difference here so this is device id or sorry device model os os version combined with ip address uh this one is ip address this one is user agent and ip address so you can see the different ways whether it's probabilistic or deterministic there's multiple ways that you can attribute sales to an impression i'll just show you a few more sessions so this looks like it was on an android phone they're on snap tube they saw a bunch of ads on snap tube and then they made a purchase on the same day that saw an impression so it looks like about five hours later and they spent 186 dollars and we know this through ipmatch so this is a a full transparent view of how each actual order id and impression served to that same consumer was achieved through that whole customer journey and then the different ways that we match this and you know we're fully transparent with our data so you can match this to your point of sale data you can download these transactions and they're all available at your fingertips next we're going to take a look at walk-ins here so walk-ins are another kpi that we track for a campaign so for a cannabis retailer not only do we want to drive e-commerce sales but we also want to drive walk-ins for retailers so depending on you know your kpis like we'll definitely want to prioritize walk-ins for the campaign so this campaign we ran uh for the course of 90 days see the investment on this campaign was three thousand dollars over 90 days and off that 3 000 investment we served 432 000 impressions that generated 809 total walk-ins with 429 unique customer walk-ins uh with a walk-in rate of just be above 0.18 percent so you can see that total walk-ins per day over time so it looks like you know we really picked up around 420 through cinco de mayo and then it kind of blended off at the end here you can see the heat map of the actual location so you can see that this is in long beach uh you can see the visits by day so it looks like the number one day from visits is first time visits is uh saturday and returning business is saturday as well uh monday also has a lot of returning visitors so we break that up by returning to new visitors you can see it the cumulative kind of overview so total number of walk-ins over time so this is very powerful powerful stuff if you're looking at a total walk-ins for your campaign and this is all done through geo-fencing your actual retail location and from that we can see how many people walked into a retail store uh from seeing one of your ads all right now let's look at another kpi we can actually try track signups for campaigns as well so let's say you have a delivery service campaign and you're

{Campaign Example: Delivery Service Campaign}

looking to just drive people to sign up for your loyalty program so this partner of ours spent twenty five thousand dollars over the course of 90 days and we were able to generate 5400 people to sign up for the loyalty program which is a cost per lead or cost per acquisition of just four dollars and 60 cents per lead so that's another kpi that we can leverage it's the same thing as the transactions so you can actually see the the ads here on the left iphones so this is the unique sign up id how many impressions were served two last impressions served on 321 and they purchased on 325 and we know this through the combination of user agent ip address same thing here it can sort by i wanted to just do like a tool change everything let's just look at 20 days so you can actually see see this over time and let's look at a brand co-op marketing campaign here so we've covered uh you know display campaigns

{Case Study: A Brand in California Looking to Drive Sale Through a Retail Partner}

really driving e-commerce sales we've talked about walk-ins we've talked about sign ups let's also talk about about a brand co-marketing campaign so this was a brand in northern california that was looking to drive sales through a retail partner so they spent 1500 over the course of 90 days so 500 a month on that campaign and they were able to generate um receive 262 000 impressions based on that seven dollar cpm and from that they were able to generate 428 clicks to the campaign with 190 transactions in 96 walk-ins with a total revenue generated of 26 547 based on that 1500 investment so this campaign took place last year so you can see the total number of clicks you can sort by click-through rate transaction count total walk-ins or observations so i'm going to show transaction count by walk-ins so here is their total transit actually let's do transaction total so transaction counts it looks like we have four transactions on this day with one walk-in and they're considered separate so if they if they made a purchase through the e-commerce menu they would not be counted as a walk-in here so you can use this to sort your performance over time and same thing you can look at the display performance is looking at the total clicks and click-through rate of the campaign you can look at the transactions in walk-ins which is all the different tabs that i already showed you observations the transaction total and let's look at impressions so we served impressions for the 90 days and then once those impressions were served we tracked sales for another 30 days because of the 30-day attribution window so that's why you'll see the impressions here in blue and then they can transactions continue to follow in teal so that's a little bit about the display campaign i'm actually going to pull up the paid search campaign next just so you have an idea of kind of what that looks like for our partners all right

{Campaign Example: Paid Search Campaign}

all right here we're looking at a an mso retailer of ours uh across like five ten states uh so we you can see that uh this is the summary page of the google ads campaign so we

{Spending Money on Marketing Channels Outside of Weedmaps}

{Allocating Budget in the Ideal Marketing Mix}

]