3 Ways You Can—and Should—Be Using Your Cannabis First-Party Data

{}Introduction

Jake Litke: I'm your host, Jake Litke, the CEO of MediaJel, and today we have Matt Taverna from Statera. Statera is a data company that— actually, you know what, Matt, I'll let you introduce Statera and your business.

Matt Taverna: yeah, um, we are a data solutions company, we build a 260 million person data set of everybody in the United States, and we then make that data actionable for advertising and outreach and analytics on kind of pieces of that data. So we have our roots in kind of state level regulation, state level politics, and we have a really good way of viewing the United States in states, and so that's how we kind of work in cannabis, using that same model of viewing the United States, that's kind of 50 states plus DC, and we work with MediaJel and other cannabis tech companies to make that data actionable and to help end users, which are companies and brands, make their own data actionable as well.

Jake Litke: great, thank you for that. The topic today is first party data, and how you can use it both in targeting and retention measurement. We should probably start off the conversation by defining what is first party data, where there's first party data, second party data, third-party data, zero party data.

First-Party Data Definition

Matt Taverna: really, we're talking about first party data today, and I'd like to define it in the context of how first party data is collected and utilized today in, let's say, the broader CPG industry, and then we get to how that—

Jake Litke: oh yeah, I mean I think that—

Matt Taverna: so just the top lines on the differences between first, second, and third party data— second and third party data are data sets that you'd buy from someone like me, consumer data that is based on different kinds of consumer behaviors that are known, made from things like when you sign up for a warranty card or your CVS card or your grocery store card. The benefit of you getting those discounts is that you can have these groceries, but then that grocery store actually has that data, and can sometimes sell that data. First party data is data that companies keep closer to the vest and collect themselves, it's data that comes from points of sale, it's data that comes from things like loyalty programs, different kinds of opt-ins for things like text messages, and it's data that's owned and operated by a company, so for lack of a better term it's like your customer list. The way it's collected, again, is through those different feeds of different software sets that companies have, and it's largely used in traditional CPG workflows to predict customer behaviors and to see what kind of inventory you should be stocking, and the trend of your business, but at a more granular level it's used primarily to keep in touch with customers.

Jake Litke: so right, you've got a couple dimensions of data there, you've got the identity of the consumer, you have maybe various types of contact information, could be phone number, email address, home address, and then you have behavioral data, which is what is the person purchasing, how often are they interacting with your retail location or delivery service. We can talk about how you can leverage that data, but before we jump over to that, I want to touch on the fact that within the CPG industry there's a healthy ecosystem of data sharing and data connections with large companies, and there's established ways that information flows back and forth. One of the things that we struggle with in the cannabis space is that the data is largely siloed, right, so you have a POS system, maybe you've got a compliance system, you've got a loyalty platform, but for the most part none of those are connected to the ecosystem that you would traditionally see a big box retailer have access to. So if you have your data siloed like that, what are the limitations that you're running into as an owner of that data?

Matt Taverna: yeah, I mean, it doesn't give you a full view of the customer, and it also—

Limitations of Owning Customer Data

Matt Taverna: leaves holes in some of the information that you're dealing with, so for instance if you have an incomplete record on your loyalty program and you are trying to maintain a relationship with your customer through a text message or an email and that isn't working, but then you have that customer on another software platform and you may have their phone number, or a different phone number for them, rather than giving up on them as a lapsed customer, it's another way you can engage with that customer, so pulling all the data into one spot makes sure you're able to pull out all the stops in customer engagement, that's number one. And number two, there are times when, because you're not recognizing that if someone gives you an email address on one piece of software and gives you a phone number on the other, if you don't make that connection you're not even getting a full view of what that customer has bought, and so their wallet size, their buying behavior with SKUs attached to it, is kind of misrepresenting that person. So if there's, my name, Matt Taverna, if I go in and buy one product one day and I'm signed in with my email address, I come in another day and I use— it's like I just come in as a regular customer but I use my phone number, making sure that those two sales are combined into a single record is really important, so you have a full view of that customer, but also a full view of that customer profile, which gets into the other use of first party data, as a predictive element.

Jake Litke: yeah, and so if you have— this is again another space where I don't think there's, or at least hasn't been, good solutions for doing data cleaning and data appending, this is something that Statera has the capability to do, I know because I've seen you do it, where you can take your customer record, let's say you've got the very simple use case of a name, you have three columns of data, you've got a name, you have a phone number, you have an email address, but then you have gaps, right, you have an email for one person, phone number for another, maybe home address. Within the CPG industry there's, like I said, established methods for going and taking that data, sending it to a data provider, getting that data filled in, but there's not a lot that really has been too available in the cannabis space, talk to me about how that usually works if I were a retailer.

Matt Taverna: yeah, I mean, I think Jake— I think the silos in cannabis are based on the fact that we're building the plane as we take off here, and I think as an industry we're very quickly catching up to best practices, but I think unfortunately what happens is that when there are established best practices,

Data Management Challenges in the Cannabis Industry

Matt Taverna: we have the legs cut out from under us, so if you look at some recent decisions from Twilio and some other software platforms where, traditionally, you would use a Twilio or another CRM as a way to combine all that data, those aren't available for cannabis brands and companies, and so having everything, having the data live in one spot, is a challenge, and I think that's one that's quickly being solved, but I think also with an industry that is, despite very recent successes— or the kind of crystal ball of success with 280E and rescheduling and some other progress made just in the past couple days on safe banking and elsewhere— despite those hopefully-down-the-road successes, it's still an industry where there's a good sense of caution that goes along with handling data and customer data, so there has to be a certain amount of trust in where you're sending those data lists, and which other vendors are able to do these things correctly for you, to make sure you see that full customer view.

Jake Litke: so those are the challenges. I think I've seen both sides of that, I've seen some businesses that are not as cautious as they should be with their first party data, and this is where we talk about PII, personally identifiable information, there are laws on what you can and can't do with it, how you're supposed to store it, those laws vary by state in terms of opt-outs and where the data collection happened and what you can do with it, so I've seen people do things like just email customer lists around with PII, which you should not do, and then I've seen a lot of people that are taking a conservative approach and just not doing anything with their data because they're concerned they don't know how much trouble they could get into, so they just play it safe and don't do anything. How do we help people

Responsible Data Management: From Collection to Security

Jake Litke: understand the middle ground of, just in broad strokes, some basic don't-do-this, and here's what you can do if you're thoughtful. And when I talk about what you can do, what are the ways you can leverage that data for your business in terms of targeting or retention or analysis? I realize that's a broad question.

Matt Taverna: yeah, no, I mean I think dealing with any vendor, whether it's a data vendor or a software vendor, knowing what that vendor is allowed or not allowed to do with your data is really important, and obviously this is best left to a lawyer, but what I usually say to a prospective customer or colleague is that you should be able to look at the terms— very clearly, in plain English, not legalese— and know what the vendor is going to do with your data. At Statera we have pretty clear rules on data destruction, that we will do a project and then your data will not stick around on our servers, we're not going to go use it for something else, we're not going to use it to create other products, we're not going to give that away to your competitors or other clients. It's kind of knowing what is possible there, and then I think using the best practices on data security, so there are pretty easy ways to securely send data files, things like Dropbox, Box.com, are very industry-standard freeware, but freeware in a way that they're well known and secure, and other data companies and vendors have their own secure way of file transfers that they've built on their own. And I think there are also just levels of security that you should be looking at, so a lot of this is common sense, but a lot of it is also based on laws and your attorney and the way that you should be looking at levels of security. So if it's one thing in particular that I really rely on our customers' attorney for— I don't need to talk about lawyers all the time here— it's if you're a medicinal dispensary, and it gets into an area of data that is kind of a true gray area for cannabis, because it does take HIPAA into consideration, but again there's a customer aspect to that, but then there's kind of the top level of security, and then there's kind of other aspects of data security that are much less scrutinized, and so things like loyalty programs are, I think, very valuable to you as a company, and I think those are data elements where you just use common sense in the way that you transfer that data to your vendor and know what they're going to do with it. You can ask your vendor for a certificate of destruction after they do a project, and have somebody walk through what they actually did with your data, and have the different software companies that you work with, if it's your point of sale company to your loyalty platform to your data vendor to programmatic media buying, have them talk to each other and be familiar with what you're trying to do, so treat this in a more holistic way and break down those silos.

Jake Litke: yeah, so a good way to think about it is like the whole journey of the data, from the beginning to the end. At the beginning you're going to collect some information, you should have a privacy policy of some sort with your customer, where you're outlining, hey, I'm going to have access to this data that you're giving me and here's what I'm going to do with it. Then when you start transporting that data to other people, to append it or to use it for targeting or something else, the other entity or vendor that you work with should have something called a data processing agreement, or something along those lines, that outlines what their rights are, because once you send information to someone you should have a clear understanding of how they're going to store that data, what they're going to do with it, do they have the ability to resell it, and what happens when the use case is over— how are they going to destroy the data or return it to you? So if you think about collecting a person's identity, if you are a retailer, you need to understand the entry point and then everywhere it can go along the lines. One of the problems with sending data files via email— it's not necessarily that the email itself is inherently insecure, it's the fact that the email sits in someone's inbox, most people use their inbox as their database, and so you've got this file floating around, it could be on someone's laptop, so you really want to think about where it's going to be at all times.

Matt Taverna: and also, if you think about an email— if you think about what it's like a Dropbox or Box.com, that's kind of multiple endpoints accessing one place, an email, when you send that, you're creating almost three different versions of the file, there's the original version, and now one in your sent box, and then one in the other person's inbox, and if they file it, it's like an exponential growth of the placement of where that data lives, that's also just problematic because, like Jake said, the wrong click of send can be catastrophic.

Jake Litke: so, okay, now that we've covered this sort of boring warning stuff of what you shouldn't do— and again, you should get legal advice, these are sort of broad descriptions we're giving here, and neither of us are lawyers as far as I know— now let's talk about what you can do with the data.

Matt Taverna: yeah, this shouldn't be scary at all, actually it should be a really great way to— so you have a view of your customer now, and if you think about the way we think about business, this is actually a really valuable asset to your company, and it's one that should be maintained as such, the maintenance of that data is super important. So say you've gone through the steps of unsiloing your software, and you have your point of sale being synced now with your loyalty platform, that's being synced with any other software you're using for text messaging and things like that, and you have all this data flowing back, and you have Matt Taverna, and he comes in and I have been the model customer, right, I've signed up for all the things and I order online and every SKU that I bought— so what can you do with that? You can create profiles of folks.

Creating Customer Profiles for Effective Marketing Strategies

Matt Taverna: I think a really good way to think about this is in the traditional CPG sense of where you've done this before, one thing— it's like a very high end, like a Bloomingdale's card, right, it's like the highest end thing, my mom had one and it was like a very special thing, we did that on Christmas time, Bloomingdale's, you know, they're tracking this stuff and they send you these magazines and this engagement and they want you to come in, it's like, oh, come in early and we're gonna open early for you, we'll open late and give you a shopper, and those are some of the things they do to entice people in, but it's not only because they know your past history, but they know that there are people like you that have done this as well. So what they've done is they're taking your data and they look at, you know, if it's Matt Taverna, a 42 year old male, they'll look at my record and they see, oh, I bought a coat last Christmas, but what do other 42 year old men buy at Bloomingdale's, and so they can compare and contrast customer records and say, well, this is what the 42 year old male profile looks like, this guy bought a coat, let's sell him a suitcase next time. So with cannabis you're doing the same thing, you can create customer profiles by looking at customer behaviors, and this is something that the cannabis data community, places like New Frontier Data, have done really well in a holistic way. Using your first party data to continue to enrich that type of data makes it invaluable, because it makes it specific to your store, so if you know what a 42 year old male does in my store, that's really important data, because it can begin to be directional on your customer outreach, so if I go and buy a three pack of pre-rolls every Friday, it's a no-brainer to send me the three pack of pre-rolls, but then you can look at what else do you think I should be buying, what else does the 42, or 35 to 45 year old male, what do they buy, let's sell them that. So you go from the downward spiral of the discount code to the upward spiral of an upsell, so hopefully I won't be buying a three pack of pre-rolls, I'll be buying more, just because you can assume certain things, because the one thing to know in human behavior that I've found with data is the best predictor of future behavior is past behavior. That not only applies to the individual, but it applies if you can create profiles that are accurate enough to make these predictions, and tapping into your first party data is a great way to do that. So having the ability to have all the contact information attached to it— address, email address, phone number— is a way to engage not only through traditional means, texting and email, but what's continuing to open up wider and wider is traditional advertising, but not like traditional digital advertising that other CPG brands use, having that contact information links you to certain digital elements like connected TV, certain apps, and certain other web platforms that are actually quite open to taking cannabis ads these days, and there are many companies, MediaJel included, that can do that work now, and there's a strong industry effort, if you look at the Cannabis Media Council, Cannabis Marketing Association, entities that have a strong interest in helping achieve the normalization of not only the cannabis media but the entire workflow, so from what you're doing with your data, making that actionable, is something that's top of mind.

Jake Litke: sure, so the other thing that allows you to do, what you're mentioning, is, okay, this is what I know about my existing customer and their behavior, here's how I can look at my own data and say this person is similar to this other person, so I should give them this offer because we see that it works with this cohort. By taking your first party data, having it organized, and then leveraging data services like Statera, or something with MediaJel, you can then look at the broader audience of people that are not yet your customers, because you have a finite number of people at market that you know who they are, but if you say, okay, I know that 42 year old males that live in this zip code or have these other demographic profiles, homeowner versus not, political affiliation— there's a whole bunch of dimensions you can measure people on, you can take your existing customers and connect that with the larger data set that's available, well established within the CPG industry, and find people in your market that have those same similar demographic traits, and advertise to them. So this is for new customer acquisition, where you can make your targeted digital advertising more effective, because you're targeting people that over-index for a specific type of product, so that allows you to say, okay, we've got a thousand customers, just round numbers, and we know that there's a cohort of them in this age group and they tend to buy three packs of pre-rolls, so we know there's an affinity for that particular product, let's build a campaign advertising that specific product to everyone in your radius that you want to be marketing to, in your location, that matches that same demographic profile, you're going to see a better return on investment when you're doing marketing that way.

Matt Taverna: and I think, if you're thinking long term, you begin to engage with folks that have a propensity to be a buyer or customer that had never been talked to before, and there's a wealth of information out there that says the cannabis industry is a little bit of a plateau right now, and part of that's because of regulation and the stifling that way, but talking to customers, talking to potential customers, and the education on how to actually go do it, go to a dispensary, can be a great way of getting people in the door. So knowing who to do that with, and not just throwing your money into the ether, having targeted communications based on your own data is the best way to start actually talking to customers, asking them to come into your store. I think sometimes we lose focus on the fact that if you're under the age of, I don't know, 25 or 35, it might be a scary experience to go into a dispensary just because when you were a kid this was all prohibited, so coming out of prohibition, knowing that targeted group of customers and having the golden nuggets in your own hands of your own customer data and being able to build those audiences and build that

Building a Data-Driven Marketing Strategy With Your First-Party Data

Matt Taverna: communication is really important, and it's a way not only to enrich your own customer base, but it grows the industry as a whole.

Jake Litke: yeah, and there's many schools of thought that count your first party data as really the only true piece of evergreen IP that you have, meaning it's an asset that you own as a business, you can keep using it in different ways to provide value for yourself without using it up, it doesn't get used up, you can continue to use it, and this is the reason why Facebook advertising frequently works so well, or has worked so well in the past, it's a little different now, but it's because Facebook knows all this information about their users. By building your own first party data set, having custody of it, enriching it, appending it, doing all the different things you can, you're in control of your own destiny a little more, because now you are the owner of that customer record and you're not reliant on Facebook or another party to reach that audience. They make it pretty easy, you can say I want to target people on Facebook that match this basic profile, this is the same tools that we're providing, but now you own the tools, you're not renting them from Facebook, and it allows you to do a lot more with it.

Matt Taverna: yeah, and I think it starts with just the ability to really begin tapping into— the more knowledge you have about your customer, the more knowledge you have about your own brand, and it's an upward spiral of success, there is that— what is really special about your dispensary that people like, and you can begin to make those observations based on a holistic view of your customer base, rather than just having the weekly discounts, is there another special aspect of your dispensary and the products you're bringing in, and the way you're presenting this stuff, that actually makes people come back, that makes a cart size a bit bigger or smaller if you're not offering a certain product that week. Being able to actually have a view of who's doing what, when, at the individual level, I think is invaluable, and then Jake talks about the value of an asset— when I worked in politics, a lot of the time you have to assign values to a lot of the things that political campaigns are working on, so one of the things that at the end of a campaign you assign value to is the data that the campaign's sitting on, and companies do this as well, CPG companies traditionally have a value assigned to their customer lists, so if you're sitting on several hundred thousand customers, even tens of thousands, that data in and of itself is a valuable asset to your company, and you can include that in your evaluation, if we want to talk about the business aspect of knowing what your business is worth, because a customer that comes in once a week, or a customer that comes in once a month and makes a big purchase, is a valuable thing to know about, and the trends you can build on top of that, with the right marketing behind it, can be invaluable.

Jake Litke: I mean, which kind of brings us full circle into how important it is to—

Matt Taverna: I think the one thing people should be most cautious about is the lack of maintenance on this type of data, and once you learn how to do the maintenance, and keep that cycle of maintenance up, it's not something you can do once in a while, it's like a marathoner can't train for a marathon by running 10 miles once a month, you need to do it every day— I don't know, I've seen some Instagram posts of people who claim they can do that.

Jake Litke: yeah, I've been talking about that, but when you say maintenance, we're talking about what, exactly?

Matt Taverna: yeah, so once you have the pie in the sky, you have a hundred thousand customers, and you have maintained it, and you're rolling and growing your customer list, what do you do then, on a daily, weekly, monthly basis— you have to look to see if people have moved, the US Postal Service has a national change of address service you can run against, and Statera offers that service, you can see if people have changed their phone number, because people change their cell phone if they move, or if they just use a prepaid phone. Maintaining that base level contact information is important, but also there are times you can just ask a customer, in an email, you don't always have to offer them a discount or a business proposition, you can ask them to update— hey, what do you like about hearing from us, can you give us your address again, do you want to hear from us still— give people the chance to engage with you at a more personal level, and that's not possible if you don't have accurate data and an updated style of data to actually make those asks.

Jake Litke: yeah, you see that commonly with many consumer apps, probably the most common one is banking apps or credit card apps, you sign in and every few months they ask you things like what is your household income, or is this still the right contact info, and that's them doing maintenance on their data, making sure they have accurate information.

Matt Taverna: yep.

Jake Litke: all right, so we do have a Q&A here in the Zoom, if anyone has questions you can type them in here and we'll do our best to answer them— again, neither of us are lawyers, so don't give us contracts to read— but happy to answer any questions anyone has, we've got probably about 10 more minutes here. Okay, so let's talk a little bit about using data to make decisions on marketing, you mentioned it's very common within the cannabis

Leveraging Data to Segment Audiences and Maximize Profit Margin

Jake Litke: industry to offer discounts as a way of getting people in the door, and it is a good tool, but it's a hammer, it definitely has an effect when you hit people with large discounts, but you can use your data to understand which of your customers you can engage with and drive retention or upsells that don't necessarily require a discount, and that's an area where you can get better margins from your existing customers. Maybe you could talk through an example of how someone would do that.

Matt Taverna: yeah, I think it goes to creating what we'd call a list segmentation, so taking your list— let's use a hundred thousand people as a nice round number— and say you have a really good way of maintaining what SKUs people are buying, and categorizing the actual spend per visit. You can categorize people a few different ways: how often they visit, how much money they spend when they visit, or what products they buy. So if it's how often they visit, you can see, if you have a date for when somebody made an online order and then came and picked it up or had delivery, you can go back to your email and see whether a purchase happened connected back to your actual marketing, and see which of those hundred thousand were affected by that discount email, and because you have the SKUs lined up to an individual, you can see pretty clearly if that person came in that day, or that week, off the discount email, or if you announce a new, more high-end, more expensive product, and you communicated that out in an email or a text or a digital campaign, you can look at a date and see that person bought one of those higher end products on that day, and attribute that back to a campaign. Then you begin to segment the data, so you can look at, hey, this is a high-end group of buyers, this is definitely a discount group of buyers, this is a group that comes in once a month and buys a pack of edibles, and it begins to build efficiencies throughout everything you're doing, because you can take your marketing and say, here's 50,000 people we shouldn't put on a discount campaign, here's 30,000 people we shouldn't try to sell beverages to because they've never bought beverages, this guy buys pre-rolls every Friday night, he's not going to come in and buy beverages one day, or he might, but we shouldn't be spending money to make that case, because that's not a good use of our budget at this time.

Jake Litke: yeah, another thing we've seen is if you are doing discounts, what is that discount— are you offering 10, 20, 30, I've seen people go all the way up to 40 depending on their goals, but understanding who you should be offering discounts to, and even what product is featured within the creative— so if you know what people tend to purchase, this cohort we know purchases pre-rolls, so let's target a set of creatives, the actual ad units themselves, that feature a picture of pre-rolls, assuming you're in a state where that's allowed, and then the next dimension is how much of a discount do we want to offer them, so you can further segment that group into people who engage when we do 25% off— there's usually a big psychological difference between 15 and 25 off, but people have different psychologies, so inside that group some will find 15 off more meaningful, and it may be because they make larger purchases, so when you do the internal math you say, okay, when I go in I'm going to spend a hundred and fifty dollars, or let's say a hundred dollars, that means 15 off, I'm gonna save 15, then you equate that to, well, basically I'm getting a whole free set of pre-rolls in my mind, whereas if you have someone who makes smaller purchases, say coming in and spending thirty dollars, fifteen percent of thirty dollars is less meaningful than thirty percent of thirty dollars. So being able to understand the data you get from your customers on what they're purchasing and what they're interacting with will allow you to make more money as a business without necessarily having to acquire new customers— obviously you want to acquire new customers too, but what I'm specifically talking about is using data to increase your margins from your existing flow of customers.

Matt Taverna: and it's also just connecting— I think one of the things that's done really well by cannabis companies is knowing the inventories and the product side, but then connecting that back to your customer base and making it truly actionable, rather than just on-stock and what's going to be in your inventory, is something that can be done with good maintenance of first party data, and connecting that back to the actual SKUs that individuals are buying, based on those inventory conversations.

Closing Remarks

Jake Litke: okay, well, lots of useful stuff here, we're wrapping up on time, do you have any last bits of advice or nuggets of information you want to share?

Matt Taverna: just a shout out to the Massachusetts cannabis market, I moved from Boston, Mass, it's one of the unsung gems of the cannabis industry here in the United States, and I'm in Maryland now, I'm gonna learn that industry as well, but it's just birthed on the adult use level, and happy to be in Maryland.

Jake Litke: so have you been into an adult use dispensary in Maryland?

Matt Taverna: no, not yet, so I'm gonna— do you know how far the nearest one is? Like 10 minutes away.

Jake Litke: all right, well, you'll have to report back and see how it compares to Boston, there's obviously massive differences in each market between what the dispensary experience is like. Great, well, thank you so much for your time, Matt, how can people reach you if they want more information or want to ask you something?

Matt Taverna: yeah, hit me up on LinkedIn if you found me there, Matt Taverna, I'm also at matt@statera.com, we'll be in touch, I assume you'll reach me on LinkedIn, I'll respond to any and all messages there, as well as my email.

Jake Litke: great, thank you so much, this is Jake Litke, CEO, MediaJel, signing off for Cannabis Marketing Live, everyone have a great weekend.

Matt Taverna: thank you Jake, bye.

{}Introduction

Jake Litke: I'm your host, Jake Litke, the CEO of MediaJel, and today we have Matt Taverna from Statera. Statera is a data company that— actually, you know what, Matt, I'll let you introduce Statera and your business.

Matt Taverna: yeah, um, we are a data solutions company, we build a 260 million person data set of everybody in the United States, and we then make that data actionable for advertising and outreach and analytics on kind of pieces of that data. So we have our roots in kind of state level regulation, state level politics, and we have a really good way of viewing the United States in states, and so that's how we kind of work in cannabis, using that same model of viewing the United States, that's kind of 50 states plus DC, and we work with MediaJel and other cannabis tech companies to make that data actionable and to help end users, which are companies and brands, make their own data actionable as well.

Jake Litke: great, thank you for that. The topic today is first party data, and how you can use it both in targeting and retention measurement. We should probably start off the conversation by defining what is first party data, where there's first party data, second party data, third-party data, zero party data.

First-Party Data Definition

Matt Taverna: really, we're talking about first party data today, and I'd like to define it in the context of how first party data is collected and utilized today in, let's say, the broader CPG industry, and then we get to how that—

Jake Litke: oh yeah, I mean I think that—

Matt Taverna: so just the top lines on the differences between first, second, and third party data— second and third party data are data sets that you'd buy from someone like me, consumer data that is based on different kinds of consumer behaviors that are known, made from things like when you sign up for a warranty card or your CVS card or your grocery store card. The benefit of you getting those discounts is that you can have these groceries, but then that grocery store actually has that data, and can sometimes sell that data. First party data is data that companies keep closer to the vest and collect themselves, it's data that comes from points of sale, it's data that comes from things like loyalty programs, different kinds of opt-ins for things like text messages, and it's data that's owned and operated by a company, so for lack of a better term it's like your customer list. The way it's collected, again, is through those different feeds of different software sets that companies have, and it's largely used in traditional CPG workflows to predict customer behaviors and to see what kind of inventory you should be stocking, and the trend of your business, but at a more granular level it's used primarily to keep in touch with customers.

Jake Litke: so right, you've got a couple dimensions of data there, you've got the identity of the consumer, you have maybe various types of contact information, could be phone number, email address, home address, and then you have behavioral data, which is what is the person purchasing, how often are they interacting with your retail location or delivery service. We can talk about how you can leverage that data, but before we jump over to that, I want to touch on the fact that within the CPG industry there's a healthy ecosystem of data sharing and data connections with large companies, and there's established ways that information flows back and forth. One of the things that we struggle with in the cannabis space is that the data is largely siloed, right, so you have a POS system, maybe you've got a compliance system, you've got a loyalty platform, but for the most part none of those are connected to the ecosystem that you would traditionally see a big box retailer have access to. So if you have your data siloed like that, what are the limitations that you're running into as an owner of that data?

Matt Taverna: yeah, I mean, it doesn't give you a full view of the customer, and it also—

Limitations of Owning Customer Data

Matt Taverna: leaves holes in some of the information that you're dealing with, so for instance if you have an incomplete record on your loyalty program and you are trying to maintain a relationship with your customer through a text message or an email and that isn't working, but then you have that customer on another software platform and you may have their phone number, or a different phone number for them, rather than giving up on them as a lapsed customer, it's another way you can engage with that customer, so pulling all the data into one spot makes sure you're able to pull out all the stops in customer engagement, that's number one. And number two, there are times when, because you're not recognizing that if someone gives you an email address on one piece of software and gives you a phone number on the other, if you don't make that connection you're not even getting a full view of what that customer has bought, and so their wallet size, their buying behavior with SKUs attached to it, is kind of misrepresenting that person. So if there's, my name, Matt Taverna, if I go in and buy one product one day and I'm signed in with my email address, I come in another day and I use— it's like I just come in as a regular customer but I use my phone number, making sure that those two sales are combined into a single record is really important, so you have a full view of that customer, but also a full view of that customer profile, which gets into the other use of first party data, as a predictive element.

Jake Litke: yeah, and so if you have— this is again another space where I don't think there's, or at least hasn't been, good solutions for doing data cleaning and data appending, this is something that Statera has the capability to do, I know because I've seen you do it, where you can take your customer record, let's say you've got the very simple use case of a name, you have three columns of data, you've got a name, you have a phone number, you have an email address, but then you have gaps, right, you have an email for one person, phone number for another, maybe home address. Within the CPG industry there's, like I said, established methods for going and taking that data, sending it to a data provider, getting that data filled in, but there's not a lot that really has been too available in the cannabis space, talk to me about how that usually works if I were a retailer.

Matt Taverna: yeah, I mean, I think Jake— I think the silos in cannabis are based on the fact that we're building the plane as we take off here, and I think as an industry we're very quickly catching up to best practices, but I think unfortunately what happens is that when there are established best practices,

Data Management Challenges in the Cannabis Industry

Matt Taverna: we have the legs cut out from under us, so if you look at some recent decisions from Twilio and some other software platforms where, traditionally, you would use a Twilio or another CRM as a way to combine all that data, those aren't available for cannabis brands and companies, and so having everything, having the data live in one spot, is a challenge, and I think that's one that's quickly being solved, but I think also with an industry that is, despite very recent successes— or the kind of crystal ball of success with 280E and rescheduling and some other progress made just in the past couple days on safe banking and elsewhere— despite those hopefully-down-the-road successes, it's still an industry where there's a good sense of caution that goes along with handling data and customer data, so there has to be a certain amount of trust in where you're sending those data lists, and which other vendors are able to do these things correctly for you, to make sure you see that full customer view.

Jake Litke: so those are the challenges. I think I've seen both sides of that, I've seen some businesses that are not as cautious as they should be with their first party data, and this is where we talk about PII, personally identifiable information, there are laws on what you can and can't do with it, how you're supposed to store it, those laws vary by state in terms of opt-outs and where the data collection happened and what you can do with it, so I've seen people do things like just email customer lists around with PII, which you should not do, and then I've seen a lot of people that are taking a conservative approach and just not doing anything with their data because they're concerned they don't know how much trouble they could get into, so they just play it safe and don't do anything. How do we help people

Responsible Data Management: From Collection to Security

Jake Litke: understand the middle ground of, just in broad strokes, some basic don't-do-this, and here's what you can do if you're thoughtful. And when I talk about what you can do, what are the ways you can leverage that data for your business in terms of targeting or retention or analysis? I realize that's a broad question.

Matt Taverna: yeah, no, I mean I think dealing with any vendor, whether it's a data vendor or a software vendor, knowing what that vendor is allowed or not allowed to do with your data is really important, and obviously this is best left to a lawyer, but what I usually say to a prospective customer or colleague is that you should be able to look at the terms— very clearly, in plain English, not legalese— and know what the vendor is going to do with your data. At Statera we have pretty clear rules on data destruction, that we will do a project and then your data will not stick around on our servers, we're not going to go use it for something else, we're not going to use it to create other products, we're not going to give that away to your competitors or other clients. It's kind of knowing what is possible there, and then I think using the best practices on data security, so there are pretty easy ways to securely send data files, things like Dropbox, Box.com, are very industry-standard freeware, but freeware in a way that they're well known and secure, and other data companies and vendors have their own secure way of file transfers that they've built on their own. And I think there are also just levels of security that you should be looking at, so a lot of this is common sense, but a lot of it is also based on laws and your attorney and the way that you should be looking at levels of security. So if it's one thing in particular that I really rely on our customers' attorney for— I don't need to talk about lawyers all the time here— it's if you're a medicinal dispensary, and it gets into an area of data that is kind of a true gray area for cannabis, because it does take HIPAA into consideration, but again there's a customer aspect to that, but then there's kind of the top level of security, and then there's kind of other aspects of data security that are much less scrutinized, and so things like loyalty programs are, I think, very valuable to you as a company, and I think those are data elements where you just use common sense in the way that you transfer that data to your vendor and know what they're going to do with it. You can ask your vendor for a certificate of destruction after they do a project, and have somebody walk through what they actually did with your data, and have the different software companies that you work with, if it's your point of sale company to your loyalty platform to your data vendor to programmatic media buying, have them talk to each other and be familiar with what you're trying to do, so treat this in a more holistic way and break down those silos.

Jake Litke: yeah, so a good way to think about it is like the whole journey of the data, from the beginning to the end. At the beginning you're going to collect some information, you should have a privacy policy of some sort with your customer, where you're outlining, hey, I'm going to have access to this data that you're giving me and here's what I'm going to do with it. Then when you start transporting that data to other people, to append it or to use it for targeting or something else, the other entity or vendor that you work with should have something called a data processing agreement, or something along those lines, that outlines what their rights are, because once you send information to someone you should have a clear understanding of how they're going to store that data, what they're going to do with it, do they have the ability to resell it, and what happens when the use case is over— how are they going to destroy the data or return it to you? So if you think about collecting a person's identity, if you are a retailer, you need to understand the entry point and then everywhere it can go along the lines. One of the problems with sending data files via email— it's not necessarily that the email itself is inherently insecure, it's the fact that the email sits in someone's inbox, most people use their inbox as their database, and so you've got this file floating around, it could be on someone's laptop, so you really want to think about where it's going to be at all times.

Matt Taverna: and also, if you think about an email— if you think about what it's like a Dropbox or Box.com, that's kind of multiple endpoints accessing one place, an email, when you send that, you're creating almost three different versions of the file, there's the original version, and now one in your sent box, and then one in the other person's inbox, and if they file it, it's like an exponential growth of the placement of where that data lives, that's also just problematic because, like Jake said, the wrong click of send can be catastrophic.

Jake Litke: so, okay, now that we've covered this sort of boring warning stuff of what you shouldn't do— and again, you should get legal advice, these are sort of broad descriptions we're giving here, and neither of us are lawyers as far as I know— now let's talk about what you can do with the data.

Matt Taverna: yeah, this shouldn't be scary at all, actually it should be a really great way to— so you have a view of your customer now, and if you think about the way we think about business, this is actually a really valuable asset to your company, and it's one that should be maintained as such, the maintenance of that data is super important. So say you've gone through the steps of unsiloing your software, and you have your point of sale being synced now with your loyalty platform, that's being synced with any other software you're using for text messaging and things like that, and you have all this data flowing back, and you have Matt Taverna, and he comes in and I have been the model customer, right, I've signed up for all the things and I order online and every SKU that I bought— so what can you do with that? You can create profiles of folks.

Creating Customer Profiles for Effective Marketing Strategies

Matt Taverna: I think a really good way to think about this is in the traditional CPG sense of where you've done this before, one thing— it's like a very high end, like a Bloomingdale's card, right, it's like the highest end thing, my mom had one and it was like a very special thing, we did that on Christmas time, Bloomingdale's, you know, they're tracking this stuff and they send you these magazines and this engagement and they want you to come in, it's like, oh, come in early and we're gonna open early for you, we'll open late and give you a shopper, and those are some of the things they do to entice people in, but it's not only because they know your past history, but they know that there are people like you that have done this as well. So what they've done is they're taking your data and they look at, you know, if it's Matt Taverna, a 42 year old male, they'll look at my record and they see, oh, I bought a coat last Christmas, but what do other 42 year old men buy at Bloomingdale's, and so they can compare and contrast customer records and say, well, this is what the 42 year old male profile looks like, this guy bought a coat, let's sell him a suitcase next time. So with cannabis you're doing the same thing, you can create customer profiles by looking at customer behaviors, and this is something that the cannabis data community, places like New Frontier Data, have done really well in a holistic way. Using your first party data to continue to enrich that type of data makes it invaluable, because it makes it specific to your store, so if you know what a 42 year old male does in my store, that's really important data, because it can begin to be directional on your customer outreach, so if I go and buy a three pack of pre-rolls every Friday, it's a no-brainer to send me the three pack of pre-rolls, but then you can look at what else do you think I should be buying, what else does the 42, or 35 to 45 year old male, what do they buy, let's sell them that. So you go from the downward spiral of the discount code to the upward spiral of an upsell, so hopefully I won't be buying a three pack of pre-rolls, I'll be buying more, just because you can assume certain things, because the one thing to know in human behavior that I've found with data is the best predictor of future behavior is past behavior. That not only applies to the individual, but it applies if you can create profiles that are accurate enough to make these predictions, and tapping into your first party data is a great way to do that. So having the ability to have all the contact information attached to it— address, email address, phone number— is a way to engage not only through traditional means, texting and email, but what's continuing to open up wider and wider is traditional advertising, but not like traditional digital advertising that other CPG brands use, having that contact information links you to certain digital elements like connected TV, certain apps, and certain other web platforms that are actually quite open to taking cannabis ads these days, and there are many companies, MediaJel included, that can do that work now, and there's a strong industry effort, if you look at the Cannabis Media Council, Cannabis Marketing Association, entities that have a strong interest in helping achieve the normalization of not only the cannabis media but the entire workflow, so from what you're doing with your data, making that actionable, is something that's top of mind.

Jake Litke: sure, so the other thing that allows you to do, what you're mentioning, is, okay, this is what I know about my existing customer and their behavior, here's how I can look at my own data and say this person is similar to this other person, so I should give them this offer because we see that it works with this cohort. By taking your first party data, having it organized, and then leveraging data services like Statera, or something with MediaJel, you can then look at the broader audience of people that are not yet your customers, because you have a finite number of people at market that you know who they are, but if you say, okay, I know that 42 year old males that live in this zip code or have these other demographic profiles, homeowner versus not, political affiliation— there's a whole bunch of dimensions you can measure people on, you can take your existing customers and connect that with the larger data set that's available, well established within the CPG industry, and find people in your market that have those same similar demographic traits, and advertise to them. So this is for new customer acquisition, where you can make your targeted digital advertising more effective, because you're targeting people that over-index for a specific type of product, so that allows you to say, okay, we've got a thousand customers, just round numbers, and we know that there's a cohort of them in this age group and they tend to buy three packs of pre-rolls, so we know there's an affinity for that particular product, let's build a campaign advertising that specific product to everyone in your radius that you want to be marketing to, in your location, that matches that same demographic profile, you're going to see a better return on investment when you're doing marketing that way.

Matt Taverna: and I think, if you're thinking long term, you begin to engage with folks that have a propensity to be a buyer or customer that had never been talked to before, and there's a wealth of information out there that says the cannabis industry is a little bit of a plateau right now, and part of that's because of regulation and the stifling that way, but talking to customers, talking to potential customers, and the education on how to actually go do it, go to a dispensary, can be a great way of getting people in the door. So knowing who to do that with, and not just throwing your money into the ether, having targeted communications based on your own data is the best way to start actually talking to customers, asking them to come into your store. I think sometimes we lose focus on the fact that if you're under the age of, I don't know, 25 or 35, it might be a scary experience to go into a dispensary just because when you were a kid this was all prohibited, so coming out of prohibition, knowing that targeted group of customers and having the golden nuggets in your own hands of your own customer data and being able to build those audiences and build that

Building a Data-Driven Marketing Strategy With Your First-Party Data

Matt Taverna: communication is really important, and it's a way not only to enrich your own customer base, but it grows the industry as a whole.

Jake Litke: yeah, and there's many schools of thought that count your first party data as really the only true piece of evergreen IP that you have, meaning it's an asset that you own as a business, you can keep using it in different ways to provide value for yourself without using it up, it doesn't get used up, you can continue to use it, and this is the reason why Facebook advertising frequently works so well, or has worked so well in the past, it's a little different now, but it's because Facebook knows all this information about their users. By building your own first party data set, having custody of it, enriching it, appending it, doing all the different things you can, you're in control of your own destiny a little more, because now you are the owner of that customer record and you're not reliant on Facebook or another party to reach that audience. They make it pretty easy, you can say I want to target people on Facebook that match this basic profile, this is the same tools that we're providing, but now you own the tools, you're not renting them from Facebook, and it allows you to do a lot more with it.

Matt Taverna: yeah, and I think it starts with just the ability to really begin tapping into— the more knowledge you have about your customer, the more knowledge you have about your own brand, and it's an upward spiral of success, there is that— what is really special about your dispensary that people like, and you can begin to make those observations based on a holistic view of your customer base, rather than just having the weekly discounts, is there another special aspect of your dispensary and the products you're bringing in, and the way you're presenting this stuff, that actually makes people come back, that makes a cart size a bit bigger or smaller if you're not offering a certain product that week. Being able to actually have a view of who's doing what, when, at the individual level, I think is invaluable, and then Jake talks about the value of an asset— when I worked in politics, a lot of the time you have to assign values to a lot of the things that political campaigns are working on, so one of the things that at the end of a campaign you assign value to is the data that the campaign's sitting on, and companies do this as well, CPG companies traditionally have a value assigned to their customer lists, so if you're sitting on several hundred thousand customers, even tens of thousands, that data in and of itself is a valuable asset to your company, and you can include that in your evaluation, if we want to talk about the business aspect of knowing what your business is worth, because a customer that comes in once a week, or a customer that comes in once a month and makes a big purchase, is a valuable thing to know about, and the trends you can build on top of that, with the right marketing behind it, can be invaluable.

Jake Litke: I mean, which kind of brings us full circle into how important it is to—

Matt Taverna: I think the one thing people should be most cautious about is the lack of maintenance on this type of data, and once you learn how to do the maintenance, and keep that cycle of maintenance up, it's not something you can do once in a while, it's like a marathoner can't train for a marathon by running 10 miles once a month, you need to do it every day— I don't know, I've seen some Instagram posts of people who claim they can do that.

Jake Litke: yeah, I've been talking about that, but when you say maintenance, we're talking about what, exactly?

Matt Taverna: yeah, so once you have the pie in the sky, you have a hundred thousand customers, and you have maintained it, and you're rolling and growing your customer list, what do you do then, on a daily, weekly, monthly basis— you have to look to see if people have moved, the US Postal Service has a national change of address service you can run against, and Statera offers that service, you can see if people have changed their phone number, because people change their cell phone if they move, or if they just use a prepaid phone. Maintaining that base level contact information is important, but also there are times you can just ask a customer, in an email, you don't always have to offer them a discount or a business proposition, you can ask them to update— hey, what do you like about hearing from us, can you give us your address again, do you want to hear from us still— give people the chance to engage with you at a more personal level, and that's not possible if you don't have accurate data and an updated style of data to actually make those asks.

Jake Litke: yeah, you see that commonly with many consumer apps, probably the most common one is banking apps or credit card apps, you sign in and every few months they ask you things like what is your household income, or is this still the right contact info, and that's them doing maintenance on their data, making sure they have accurate information.

Matt Taverna: yep.

Jake Litke: all right, so we do have a Q&A here in the Zoom, if anyone has questions you can type them in here and we'll do our best to answer them— again, neither of us are lawyers, so don't give us contracts to read— but happy to answer any questions anyone has, we've got probably about 10 more minutes here. Okay, so let's talk a little bit about using data to make decisions on marketing, you mentioned it's very common within the cannabis

Leveraging Data to Segment Audiences and Maximize Profit Margin

Jake Litke: industry to offer discounts as a way of getting people in the door, and it is a good tool, but it's a hammer, it definitely has an effect when you hit people with large discounts, but you can use your data to understand which of your customers you can engage with and drive retention or upsells that don't necessarily require a discount, and that's an area where you can get better margins from your existing customers. Maybe you could talk through an example of how someone would do that.

Matt Taverna: yeah, I think it goes to creating what we'd call a list segmentation, so taking your list— let's use a hundred thousand people as a nice round number— and say you have a really good way of maintaining what SKUs people are buying, and categorizing the actual spend per visit. You can categorize people a few different ways: how often they visit, how much money they spend when they visit, or what products they buy. So if it's how often they visit, you can see, if you have a date for when somebody made an online order and then came and picked it up or had delivery, you can go back to your email and see whether a purchase happened connected back to your actual marketing, and see which of those hundred thousand were affected by that discount email, and because you have the SKUs lined up to an individual, you can see pretty clearly if that person came in that day, or that week, off the discount email, or if you announce a new, more high-end, more expensive product, and you communicated that out in an email or a text or a digital campaign, you can look at a date and see that person bought one of those higher end products on that day, and attribute that back to a campaign. Then you begin to segment the data, so you can look at, hey, this is a high-end group of buyers, this is definitely a discount group of buyers, this is a group that comes in once a month and buys a pack of edibles, and it begins to build efficiencies throughout everything you're doing, because you can take your marketing and say, here's 50,000 people we shouldn't put on a discount campaign, here's 30,000 people we shouldn't try to sell beverages to because they've never bought beverages, this guy buys pre-rolls every Friday night, he's not going to come in and buy beverages one day, or he might, but we shouldn't be spending money to make that case, because that's not a good use of our budget at this time.

Jake Litke: yeah, another thing we've seen is if you are doing discounts, what is that discount— are you offering 10, 20, 30, I've seen people go all the way up to 40 depending on their goals, but understanding who you should be offering discounts to, and even what product is featured within the creative— so if you know what people tend to purchase, this cohort we know purchases pre-rolls, so let's target a set of creatives, the actual ad units themselves, that feature a picture of pre-rolls, assuming you're in a state where that's allowed, and then the next dimension is how much of a discount do we want to offer them, so you can further segment that group into people who engage when we do 25% off— there's usually a big psychological difference between 15 and 25 off, but people have different psychologies, so inside that group some will find 15 off more meaningful, and it may be because they make larger purchases, so when you do the internal math you say, okay, when I go in I'm going to spend a hundred and fifty dollars, or let's say a hundred dollars, that means 15 off, I'm gonna save 15, then you equate that to, well, basically I'm getting a whole free set of pre-rolls in my mind, whereas if you have someone who makes smaller purchases, say coming in and spending thirty dollars, fifteen percent of thirty dollars is less meaningful than thirty percent of thirty dollars. So being able to understand the data you get from your customers on what they're purchasing and what they're interacting with will allow you to make more money as a business without necessarily having to acquire new customers— obviously you want to acquire new customers too, but what I'm specifically talking about is using data to increase your margins from your existing flow of customers.

Matt Taverna: and it's also just connecting— I think one of the things that's done really well by cannabis companies is knowing the inventories and the product side, but then connecting that back to your customer base and making it truly actionable, rather than just on-stock and what's going to be in your inventory, is something that can be done with good maintenance of first party data, and connecting that back to the actual SKUs that individuals are buying, based on those inventory conversations.

Closing Remarks

Jake Litke: okay, well, lots of useful stuff here, we're wrapping up on time, do you have any last bits of advice or nuggets of information you want to share?

Matt Taverna: just a shout out to the Massachusetts cannabis market, I moved from Boston, Mass, it's one of the unsung gems of the cannabis industry here in the United States, and I'm in Maryland now, I'm gonna learn that industry as well, but it's just birthed on the adult use level, and happy to be in Maryland.

Jake Litke: so have you been into an adult use dispensary in Maryland?

Matt Taverna: no, not yet, so I'm gonna— do you know how far the nearest one is? Like 10 minutes away.

Jake Litke: all right, well, you'll have to report back and see how it compares to Boston, there's obviously massive differences in each market between what the dispensary experience is like. Great, well, thank you so much for your time, Matt, how can people reach you if they want more information or want to ask you something?

Matt Taverna: yeah, hit me up on LinkedIn if you found me there, Matt Taverna, I'm also at matt@statera.com, we'll be in touch, I assume you'll reach me on LinkedIn, I'll respond to any and all messages there, as well as my email.

Jake Litke: great, thank you so much, this is Jake Litke, CEO, MediaJel, signing off for Cannabis Marketing Live, everyone have a great weekend.

Matt Taverna: thank you Jake, bye.

{}Introduction

Jake Litke: I'm your host, Jake Litke, the CEO of MediaJel, and today we have Matt Taverna from Statera. Statera is a data company that— actually, you know what, Matt, I'll let you introduce Statera and your business.

Matt Taverna: yeah, um, we are a data solutions company, we build a 260 million person data set of everybody in the United States, and we then make that data actionable for advertising and outreach and analytics on kind of pieces of that data. So we have our roots in kind of state level regulation, state level politics, and we have a really good way of viewing the United States in states, and so that's how we kind of work in cannabis, using that same model of viewing the United States, that's kind of 50 states plus DC, and we work with MediaJel and other cannabis tech companies to make that data actionable and to help end users, which are companies and brands, make their own data actionable as well.

Jake Litke: great, thank you for that. The topic today is first party data, and how you can use it both in targeting and retention measurement. We should probably start off the conversation by defining what is first party data, where there's first party data, second party data, third-party data, zero party data.

First-Party Data Definition

Matt Taverna: really, we're talking about first party data today, and I'd like to define it in the context of how first party data is collected and utilized today in, let's say, the broader CPG industry, and then we get to how that—

Jake Litke: oh yeah, I mean I think that—

Matt Taverna: so just the top lines on the differences between first, second, and third party data— second and third party data are data sets that you'd buy from someone like me, consumer data that is based on different kinds of consumer behaviors that are known, made from things like when you sign up for a warranty card or your CVS card or your grocery store card. The benefit of you getting those discounts is that you can have these groceries, but then that grocery store actually has that data, and can sometimes sell that data. First party data is data that companies keep closer to the vest and collect themselves, it's data that comes from points of sale, it's data that comes from things like loyalty programs, different kinds of opt-ins for things like text messages, and it's data that's owned and operated by a company, so for lack of a better term it's like your customer list. The way it's collected, again, is through those different feeds of different software sets that companies have, and it's largely used in traditional CPG workflows to predict customer behaviors and to see what kind of inventory you should be stocking, and the trend of your business, but at a more granular level it's used primarily to keep in touch with customers.

Jake Litke: so right, you've got a couple dimensions of data there, you've got the identity of the consumer, you have maybe various types of contact information, could be phone number, email address, home address, and then you have behavioral data, which is what is the person purchasing, how often are they interacting with your retail location or delivery service. We can talk about how you can leverage that data, but before we jump over to that, I want to touch on the fact that within the CPG industry there's a healthy ecosystem of data sharing and data connections with large companies, and there's established ways that information flows back and forth. One of the things that we struggle with in the cannabis space is that the data is largely siloed, right, so you have a POS system, maybe you've got a compliance system, you've got a loyalty platform, but for the most part none of those are connected to the ecosystem that you would traditionally see a big box retailer have access to. So if you have your data siloed like that, what are the limitations that you're running into as an owner of that data?

Matt Taverna: yeah, I mean, it doesn't give you a full view of the customer, and it also—

Limitations of Owning Customer Data

Matt Taverna: leaves holes in some of the information that you're dealing with, so for instance if you have an incomplete record on your loyalty program and you are trying to maintain a relationship with your customer through a text message or an email and that isn't working, but then you have that customer on another software platform and you may have their phone number, or a different phone number for them, rather than giving up on them as a lapsed customer, it's another way you can engage with that customer, so pulling all the data into one spot makes sure you're able to pull out all the stops in customer engagement, that's number one. And number two, there are times when, because you're not recognizing that if someone gives you an email address on one piece of software and gives you a phone number on the other, if you don't make that connection you're not even getting a full view of what that customer has bought, and so their wallet size, their buying behavior with SKUs attached to it, is kind of misrepresenting that person. So if there's, my name, Matt Taverna, if I go in and buy one product one day and I'm signed in with my email address, I come in another day and I use— it's like I just come in as a regular customer but I use my phone number, making sure that those two sales are combined into a single record is really important, so you have a full view of that customer, but also a full view of that customer profile, which gets into the other use of first party data, as a predictive element.

Jake Litke: yeah, and so if you have— this is again another space where I don't think there's, or at least hasn't been, good solutions for doing data cleaning and data appending, this is something that Statera has the capability to do, I know because I've seen you do it, where you can take your customer record, let's say you've got the very simple use case of a name, you have three columns of data, you've got a name, you have a phone number, you have an email address, but then you have gaps, right, you have an email for one person, phone number for another, maybe home address. Within the CPG industry there's, like I said, established methods for going and taking that data, sending it to a data provider, getting that data filled in, but there's not a lot that really has been too available in the cannabis space, talk to me about how that usually works if I were a retailer.

Matt Taverna: yeah, I mean, I think Jake— I think the silos in cannabis are based on the fact that we're building the plane as we take off here, and I think as an industry we're very quickly catching up to best practices, but I think unfortunately what happens is that when there are established best practices,

Data Management Challenges in the Cannabis Industry

Matt Taverna: we have the legs cut out from under us, so if you look at some recent decisions from Twilio and some other software platforms where, traditionally, you would use a Twilio or another CRM as a way to combine all that data, those aren't available for cannabis brands and companies, and so having everything, having the data live in one spot, is a challenge, and I think that's one that's quickly being solved, but I think also with an industry that is, despite very recent successes— or the kind of crystal ball of success with 280E and rescheduling and some other progress made just in the past couple days on safe banking and elsewhere— despite those hopefully-down-the-road successes, it's still an industry where there's a good sense of caution that goes along with handling data and customer data, so there has to be a certain amount of trust in where you're sending those data lists, and which other vendors are able to do these things correctly for you, to make sure you see that full customer view.

Jake Litke: so those are the challenges. I think I've seen both sides of that, I've seen some businesses that are not as cautious as they should be with their first party data, and this is where we talk about PII, personally identifiable information, there are laws on what you can and can't do with it, how you're supposed to store it, those laws vary by state in terms of opt-outs and where the data collection happened and what you can do with it, so I've seen people do things like just email customer lists around with PII, which you should not do, and then I've seen a lot of people that are taking a conservative approach and just not doing anything with their data because they're concerned they don't know how much trouble they could get into, so they just play it safe and don't do anything. How do we help people

Responsible Data Management: From Collection to Security

Jake Litke: understand the middle ground of, just in broad strokes, some basic don't-do-this, and here's what you can do if you're thoughtful. And when I talk about what you can do, what are the ways you can leverage that data for your business in terms of targeting or retention or analysis? I realize that's a broad question.

Matt Taverna: yeah, no, I mean I think dealing with any vendor, whether it's a data vendor or a software vendor, knowing what that vendor is allowed or not allowed to do with your data is really important, and obviously this is best left to a lawyer, but what I usually say to a prospective customer or colleague is that you should be able to look at the terms— very clearly, in plain English, not legalese— and know what the vendor is going to do with your data. At Statera we have pretty clear rules on data destruction, that we will do a project and then your data will not stick around on our servers, we're not going to go use it for something else, we're not going to use it to create other products, we're not going to give that away to your competitors or other clients. It's kind of knowing what is possible there, and then I think using the best practices on data security, so there are pretty easy ways to securely send data files, things like Dropbox, Box.com, are very industry-standard freeware, but freeware in a way that they're well known and secure, and other data companies and vendors have their own secure way of file transfers that they've built on their own. And I think there are also just levels of security that you should be looking at, so a lot of this is common sense, but a lot of it is also based on laws and your attorney and the way that you should be looking at levels of security. So if it's one thing in particular that I really rely on our customers' attorney for— I don't need to talk about lawyers all the time here— it's if you're a medicinal dispensary, and it gets into an area of data that is kind of a true gray area for cannabis, because it does take HIPAA into consideration, but again there's a customer aspect to that, but then there's kind of the top level of security, and then there's kind of other aspects of data security that are much less scrutinized, and so things like loyalty programs are, I think, very valuable to you as a company, and I think those are data elements where you just use common sense in the way that you transfer that data to your vendor and know what they're going to do with it. You can ask your vendor for a certificate of destruction after they do a project, and have somebody walk through what they actually did with your data, and have the different software companies that you work with, if it's your point of sale company to your loyalty platform to your data vendor to programmatic media buying, have them talk to each other and be familiar with what you're trying to do, so treat this in a more holistic way and break down those silos.

Jake Litke: yeah, so a good way to think about it is like the whole journey of the data, from the beginning to the end. At the beginning you're going to collect some information, you should have a privacy policy of some sort with your customer, where you're outlining, hey, I'm going to have access to this data that you're giving me and here's what I'm going to do with it. Then when you start transporting that data to other people, to append it or to use it for targeting or something else, the other entity or vendor that you work with should have something called a data processing agreement, or something along those lines, that outlines what their rights are, because once you send information to someone you should have a clear understanding of how they're going to store that data, what they're going to do with it, do they have the ability to resell it, and what happens when the use case is over— how are they going to destroy the data or return it to you? So if you think about collecting a person's identity, if you are a retailer, you need to understand the entry point and then everywhere it can go along the lines. One of the problems with sending data files via email— it's not necessarily that the email itself is inherently insecure, it's the fact that the email sits in someone's inbox, most people use their inbox as their database, and so you've got this file floating around, it could be on someone's laptop, so you really want to think about where it's going to be at all times.

Matt Taverna: and also, if you think about an email— if you think about what it's like a Dropbox or Box.com, that's kind of multiple endpoints accessing one place, an email, when you send that, you're creating almost three different versions of the file, there's the original version, and now one in your sent box, and then one in the other person's inbox, and if they file it, it's like an exponential growth of the placement of where that data lives, that's also just problematic because, like Jake said, the wrong click of send can be catastrophic.

Jake Litke: so, okay, now that we've covered this sort of boring warning stuff of what you shouldn't do— and again, you should get legal advice, these are sort of broad descriptions we're giving here, and neither of us are lawyers as far as I know— now let's talk about what you can do with the data.

Matt Taverna: yeah, this shouldn't be scary at all, actually it should be a really great way to— so you have a view of your customer now, and if you think about the way we think about business, this is actually a really valuable asset to your company, and it's one that should be maintained as such, the maintenance of that data is super important. So say you've gone through the steps of unsiloing your software, and you have your point of sale being synced now with your loyalty platform, that's being synced with any other software you're using for text messaging and things like that, and you have all this data flowing back, and you have Matt Taverna, and he comes in and I have been the model customer, right, I've signed up for all the things and I order online and every SKU that I bought— so what can you do with that? You can create profiles of folks.

Creating Customer Profiles for Effective Marketing Strategies

Matt Taverna: I think a really good way to think about this is in the traditional CPG sense of where you've done this before, one thing— it's like a very high end, like a Bloomingdale's card, right, it's like the highest end thing, my mom had one and it was like a very special thing, we did that on Christmas time, Bloomingdale's, you know, they're tracking this stuff and they send you these magazines and this engagement and they want you to come in, it's like, oh, come in early and we're gonna open early for you, we'll open late and give you a shopper, and those are some of the things they do to entice people in, but it's not only because they know your past history, but they know that there are people like you that have done this as well. So what they've done is they're taking your data and they look at, you know, if it's Matt Taverna, a 42 year old male, they'll look at my record and they see, oh, I bought a coat last Christmas, but what do other 42 year old men buy at Bloomingdale's, and so they can compare and contrast customer records and say, well, this is what the 42 year old male profile looks like, this guy bought a coat, let's sell him a suitcase next time. So with cannabis you're doing the same thing, you can create customer profiles by looking at customer behaviors, and this is something that the cannabis data community, places like New Frontier Data, have done really well in a holistic way. Using your first party data to continue to enrich that type of data makes it invaluable, because it makes it specific to your store, so if you know what a 42 year old male does in my store, that's really important data, because it can begin to be directional on your customer outreach, so if I go and buy a three pack of pre-rolls every Friday, it's a no-brainer to send me the three pack of pre-rolls, but then you can look at what else do you think I should be buying, what else does the 42, or 35 to 45 year old male, what do they buy, let's sell them that. So you go from the downward spiral of the discount code to the upward spiral of an upsell, so hopefully I won't be buying a three pack of pre-rolls, I'll be buying more, just because you can assume certain things, because the one thing to know in human behavior that I've found with data is the best predictor of future behavior is past behavior. That not only applies to the individual, but it applies if you can create profiles that are accurate enough to make these predictions, and tapping into your first party data is a great way to do that. So having the ability to have all the contact information attached to it— address, email address, phone number— is a way to engage not only through traditional means, texting and email, but what's continuing to open up wider and wider is traditional advertising, but not like traditional digital advertising that other CPG brands use, having that contact information links you to certain digital elements like connected TV, certain apps, and certain other web platforms that are actually quite open to taking cannabis ads these days, and there are many companies, MediaJel included, that can do that work now, and there's a strong industry effort, if you look at the Cannabis Media Council, Cannabis Marketing Association, entities that have a strong interest in helping achieve the normalization of not only the cannabis media but the entire workflow, so from what you're doing with your data, making that actionable, is something that's top of mind.

Jake Litke: sure, so the other thing that allows you to do, what you're mentioning, is, okay, this is what I know about my existing customer and their behavior, here's how I can look at my own data and say this person is similar to this other person, so I should give them this offer because we see that it works with this cohort. By taking your first party data, having it organized, and then leveraging data services like Statera, or something with MediaJel, you can then look at the broader audience of people that are not yet your customers, because you have a finite number of people at market that you know who they are, but if you say, okay, I know that 42 year old males that live in this zip code or have these other demographic profiles, homeowner versus not, political affiliation— there's a whole bunch of dimensions you can measure people on, you can take your existing customers and connect that with the larger data set that's available, well established within the CPG industry, and find people in your market that have those same similar demographic traits, and advertise to them. So this is for new customer acquisition, where you can make your targeted digital advertising more effective, because you're targeting people that over-index for a specific type of product, so that allows you to say, okay, we've got a thousand customers, just round numbers, and we know that there's a cohort of them in this age group and they tend to buy three packs of pre-rolls, so we know there's an affinity for that particular product, let's build a campaign advertising that specific product to everyone in your radius that you want to be marketing to, in your location, that matches that same demographic profile, you're going to see a better return on investment when you're doing marketing that way.

Matt Taverna: and I think, if you're thinking long term, you begin to engage with folks that have a propensity to be a buyer or customer that had never been talked to before, and there's a wealth of information out there that says the cannabis industry is a little bit of a plateau right now, and part of that's because of regulation and the stifling that way, but talking to customers, talking to potential customers, and the education on how to actually go do it, go to a dispensary, can be a great way of getting people in the door. So knowing who to do that with, and not just throwing your money into the ether, having targeted communications based on your own data is the best way to start actually talking to customers, asking them to come into your store. I think sometimes we lose focus on the fact that if you're under the age of, I don't know, 25 or 35, it might be a scary experience to go into a dispensary just because when you were a kid this was all prohibited, so coming out of prohibition, knowing that targeted group of customers and having the golden nuggets in your own hands of your own customer data and being able to build those audiences and build that

Building a Data-Driven Marketing Strategy With Your First-Party Data

Matt Taverna: communication is really important, and it's a way not only to enrich your own customer base, but it grows the industry as a whole.

Jake Litke: yeah, and there's many schools of thought that count your first party data as really the only true piece of evergreen IP that you have, meaning it's an asset that you own as a business, you can keep using it in different ways to provide value for yourself without using it up, it doesn't get used up, you can continue to use it, and this is the reason why Facebook advertising frequently works so well, or has worked so well in the past, it's a little different now, but it's because Facebook knows all this information about their users. By building your own first party data set, having custody of it, enriching it, appending it, doing all the different things you can, you're in control of your own destiny a little more, because now you are the owner of that customer record and you're not reliant on Facebook or another party to reach that audience. They make it pretty easy, you can say I want to target people on Facebook that match this basic profile, this is the same tools that we're providing, but now you own the tools, you're not renting them from Facebook, and it allows you to do a lot more with it.

Matt Taverna: yeah, and I think it starts with just the ability to really begin tapping into— the more knowledge you have about your customer, the more knowledge you have about your own brand, and it's an upward spiral of success, there is that— what is really special about your dispensary that people like, and you can begin to make those observations based on a holistic view of your customer base, rather than just having the weekly discounts, is there another special aspect of your dispensary and the products you're bringing in, and the way you're presenting this stuff, that actually makes people come back, that makes a cart size a bit bigger or smaller if you're not offering a certain product that week. Being able to actually have a view of who's doing what, when, at the individual level, I think is invaluable, and then Jake talks about the value of an asset— when I worked in politics, a lot of the time you have to assign values to a lot of the things that political campaigns are working on, so one of the things that at the end of a campaign you assign value to is the data that the campaign's sitting on, and companies do this as well, CPG companies traditionally have a value assigned to their customer lists, so if you're sitting on several hundred thousand customers, even tens of thousands, that data in and of itself is a valuable asset to your company, and you can include that in your evaluation, if we want to talk about the business aspect of knowing what your business is worth, because a customer that comes in once a week, or a customer that comes in once a month and makes a big purchase, is a valuable thing to know about, and the trends you can build on top of that, with the right marketing behind it, can be invaluable.

Jake Litke: I mean, which kind of brings us full circle into how important it is to—

Matt Taverna: I think the one thing people should be most cautious about is the lack of maintenance on this type of data, and once you learn how to do the maintenance, and keep that cycle of maintenance up, it's not something you can do once in a while, it's like a marathoner can't train for a marathon by running 10 miles once a month, you need to do it every day— I don't know, I've seen some Instagram posts of people who claim they can do that.

Jake Litke: yeah, I've been talking about that, but when you say maintenance, we're talking about what, exactly?

Matt Taverna: yeah, so once you have the pie in the sky, you have a hundred thousand customers, and you have maintained it, and you're rolling and growing your customer list, what do you do then, on a daily, weekly, monthly basis— you have to look to see if people have moved, the US Postal Service has a national change of address service you can run against, and Statera offers that service, you can see if people have changed their phone number, because people change their cell phone if they move, or if they just use a prepaid phone. Maintaining that base level contact information is important, but also there are times you can just ask a customer, in an email, you don't always have to offer them a discount or a business proposition, you can ask them to update— hey, what do you like about hearing from us, can you give us your address again, do you want to hear from us still— give people the chance to engage with you at a more personal level, and that's not possible if you don't have accurate data and an updated style of data to actually make those asks.

Jake Litke: yeah, you see that commonly with many consumer apps, probably the most common one is banking apps or credit card apps, you sign in and every few months they ask you things like what is your household income, or is this still the right contact info, and that's them doing maintenance on their data, making sure they have accurate information.

Matt Taverna: yep.

Jake Litke: all right, so we do have a Q&A here in the Zoom, if anyone has questions you can type them in here and we'll do our best to answer them— again, neither of us are lawyers, so don't give us contracts to read— but happy to answer any questions anyone has, we've got probably about 10 more minutes here. Okay, so let's talk a little bit about using data to make decisions on marketing, you mentioned it's very common within the cannabis

Leveraging Data to Segment Audiences and Maximize Profit Margin

Jake Litke: industry to offer discounts as a way of getting people in the door, and it is a good tool, but it's a hammer, it definitely has an effect when you hit people with large discounts, but you can use your data to understand which of your customers you can engage with and drive retention or upsells that don't necessarily require a discount, and that's an area where you can get better margins from your existing customers. Maybe you could talk through an example of how someone would do that.

Matt Taverna: yeah, I think it goes to creating what we'd call a list segmentation, so taking your list— let's use a hundred thousand people as a nice round number— and say you have a really good way of maintaining what SKUs people are buying, and categorizing the actual spend per visit. You can categorize people a few different ways: how often they visit, how much money they spend when they visit, or what products they buy. So if it's how often they visit, you can see, if you have a date for when somebody made an online order and then came and picked it up or had delivery, you can go back to your email and see whether a purchase happened connected back to your actual marketing, and see which of those hundred thousand were affected by that discount email, and because you have the SKUs lined up to an individual, you can see pretty clearly if that person came in that day, or that week, off the discount email, or if you announce a new, more high-end, more expensive product, and you communicated that out in an email or a text or a digital campaign, you can look at a date and see that person bought one of those higher end products on that day, and attribute that back to a campaign. Then you begin to segment the data, so you can look at, hey, this is a high-end group of buyers, this is definitely a discount group of buyers, this is a group that comes in once a month and buys a pack of edibles, and it begins to build efficiencies throughout everything you're doing, because you can take your marketing and say, here's 50,000 people we shouldn't put on a discount campaign, here's 30,000 people we shouldn't try to sell beverages to because they've never bought beverages, this guy buys pre-rolls every Friday night, he's not going to come in and buy beverages one day, or he might, but we shouldn't be spending money to make that case, because that's not a good use of our budget at this time.

Jake Litke: yeah, another thing we've seen is if you are doing discounts, what is that discount— are you offering 10, 20, 30, I've seen people go all the way up to 40 depending on their goals, but understanding who you should be offering discounts to, and even what product is featured within the creative— so if you know what people tend to purchase, this cohort we know purchases pre-rolls, so let's target a set of creatives, the actual ad units themselves, that feature a picture of pre-rolls, assuming you're in a state where that's allowed, and then the next dimension is how much of a discount do we want to offer them, so you can further segment that group into people who engage when we do 25% off— there's usually a big psychological difference between 15 and 25 off, but people have different psychologies, so inside that group some will find 15 off more meaningful, and it may be because they make larger purchases, so when you do the internal math you say, okay, when I go in I'm going to spend a hundred and fifty dollars, or let's say a hundred dollars, that means 15 off, I'm gonna save 15, then you equate that to, well, basically I'm getting a whole free set of pre-rolls in my mind, whereas if you have someone who makes smaller purchases, say coming in and spending thirty dollars, fifteen percent of thirty dollars is less meaningful than thirty percent of thirty dollars. So being able to understand the data you get from your customers on what they're purchasing and what they're interacting with will allow you to make more money as a business without necessarily having to acquire new customers— obviously you want to acquire new customers too, but what I'm specifically talking about is using data to increase your margins from your existing flow of customers.

Matt Taverna: and it's also just connecting— I think one of the things that's done really well by cannabis companies is knowing the inventories and the product side, but then connecting that back to your customer base and making it truly actionable, rather than just on-stock and what's going to be in your inventory, is something that can be done with good maintenance of first party data, and connecting that back to the actual SKUs that individuals are buying, based on those inventory conversations.

Closing Remarks

Jake Litke: okay, well, lots of useful stuff here, we're wrapping up on time, do you have any last bits of advice or nuggets of information you want to share?

Matt Taverna: just a shout out to the Massachusetts cannabis market, I moved from Boston, Mass, it's one of the unsung gems of the cannabis industry here in the United States, and I'm in Maryland now, I'm gonna learn that industry as well, but it's just birthed on the adult use level, and happy to be in Maryland.

Jake Litke: so have you been into an adult use dispensary in Maryland?

Matt Taverna: no, not yet, so I'm gonna— do you know how far the nearest one is? Like 10 minutes away.

Jake Litke: all right, well, you'll have to report back and see how it compares to Boston, there's obviously massive differences in each market between what the dispensary experience is like. Great, well, thank you so much for your time, Matt, how can people reach you if they want more information or want to ask you something?

Matt Taverna: yeah, hit me up on LinkedIn if you found me there, Matt Taverna, I'm also at matt@statera.com, we'll be in touch, I assume you'll reach me on LinkedIn, I'll respond to any and all messages there, as well as my email.

Jake Litke: great, thank you so much, this is Jake Litke, CEO, MediaJel, signing off for Cannabis Marketing Live, everyone have a great weekend.

Matt Taverna: thank you Jake, bye.