Subscribe
Share
Share
Embed
If you are seeking new ways to increase your ROI on marketing with your commerce platform, or you may be an entrepreneur who wants to grow your team and be more efficient with your online business.
Talk Commerce with Brent W. Peterson draws stories from merchants, marketers, and entrepreneurs who share their experiences in the trenches to help you learn what works and what may not in your business.
Keep up with the current news on commerce platforms, marketing trends, and what is new in the entrepreneurial world. Episodes drop every Tuesday with the occasional bonus episodes.
You can check out our daily blog post and signup for our newsletter here https://talk-commerce.com
Brent Peterson (00:02.902)
All right, welcome to this episode, this e -tail episode of Talk Commerce. Today I have Cary Lawrence. She is the CEO of Docile. Cary? Decile. Decile, I'm sorry, not docile. It's I'm feeling a little lethargic, so I said docile. Sorry about that. Tell us your day -to -day role and maybe one of your passions in life. Sure, great. And so nice to be chatting with you outside. So I am the CEO of Docile, which is basically a customer data.
and marketing platform and built by marketers. So a lot of us have those marketing backgrounds and specifically for marketers. So we can really understand how to kind of leverage that data. And as far as passions in life, I think if I were to sum it up with one word, I would say food, both cooking, which I really enjoy, as well as going out to restaurants and try new things. That's awesome. And of course, we tried to find a nice quiet place and now the food's coming out. So.
You will hear some engine noise or some plane noise and things like that. OK, so Desi, you tell us tell us what the platform does. Give us your 30 second elevator. Not a pitch, but just kind of give us a background of what it does. Sure. And our goal really is to help marketers leverage what we think is their most valuable marketing asset, which is that first party customer data. So we're kind of all the great things that you would get with a traditional customer data platform or CDP, as it's commonly known, and also an advanced analytics platform.
So we're helping these marketers to build and retain a more loyal customer base to really drive personalized marketing campaigns by use of our AI generated personas that look at both purchase data as well as those softer metrics like demographics, psychographics. And ultimately our goal is to really help them find those customers who are gonna be high value repeat purchasers, the ones that are gonna allow them to drive that profitable growth. Are you seeing...
the upset of the market where the cookie's going away and obviously iOS 14, it took all those things. Is it harder for you or is it now becoming more of a level playing field? It's actually great for us. I feel like we were very well positioned to be very relevant for a world that is moving towards more identity -based marketing and that is how we built our platform.
Brent Peterson (02:20.878)
from the ground up. So everything that we do within Desyle is based on individual known users or known customers. We're not matching on device IDs or cookies. So we've always been playing in the first party world. And I think that even though it's been like the little boy crying wolf with everyone talking about cookies going away, the time is really happening now. I think you'll sense that even at this conference. A lot of people worried about, with Chrome in particular, being like the dominant browser, just testing.
kind of 1 % of users without cookies, but it seems like Q3 of 2024 is when it's really gonna happen. So, Decile is very well positioned to help our clients navigate this new world. Some of the personalization can feel a little bit impersonal when you're using AI, at least when the first versions of Chat GPT came out, it wasn't always that great sounding.
How do you ensure that the personalization is the merchant's voice? Yeah, well for us, our models that leverage, it's really leveraging machines to look at data. So looking at all those different attributes of someone. So for you, Brent, we might know that you like to travel to Hawaii, you like to have flowered shirts and apparel, and we also know that you tend to be a repeat purchaser for...
you know, the surf brands. So we have all those little insights about you. So it's less about kind of the sort of the chat function or the language learning models and more just using machines to predict, you know, what is going to be your first, you know, after your first purchase, what is going to be the next purchase? What is going be the one after that? Are you going to be a high value customer? You know, so are you going to be someone who's a high lifetime value? So for us, it's really important when you think about those like high, um,
purchase season. So for marketers in the holiday season, for instance, it's very common that customers are getting hit with marketing messages, you know, two or three emails a day and, you know, two or three texts, you know, 2022 was the big year of SMS marketing. And so you don't want to just be blasted with like, Hey, Brent, come, come check out our store, come buy this product. What you want is a really personalized message that says, Hey, Brent, we know you tend to buy these flower button downs. So those are now in sale. And we also know that,
Brent Peterson (04:45.262)
you won't ever buy from us unless it's at least 20 % discount. So now here's your 20 % discount code and really give that equal value exchange. So really it's like we feel that customers will willingly share their email or share their information if they know that they're going to get value in return. And do you anonymize some of that to help understand what I would buy compared to what another white, bold male might buy? Yes, of course. And everything's kind of done at the audience level.
So it's really about having a platform that can create these segments very seamlessly. What's nice about our platform is we're also enriching that first party data with third party data, like demographics, psychographic, interest, and behavior. So you can create a very highly customized and valuable audience. So it could be, as I said, it could be your top 10 % highest LTV. It could be just people who buy.
you know, who tend to buy shirts. It could be men versus women. So you want to make sure that you're kind of creating those cuts. But it's not just you, Brent, we're looking at an audience of those folks who have those behaviors. I had an interesting conversation yesterday about AOV and that the average order value in somebody's cart is or somebody's for a merchant is the average of everybody, right? But then that also means that you're missing out on half of your people, maybe.
Do you help to then target more better target some of those people so you're not so average? Yeah, and actually you bring up a good point, which is one of the things and this is actually interesting. Our name decile is this is kind of where it comes from. If you think of a percentiles, you divide something into 100 parts. If you divide something to 10 parts, those are deciles. And what we found is that across the board, the top decile, that top 10 percent of customers are responsible for.
40 to 60 % of all revenue and profit for a company. So it's really important to be able to kind of find those people who are going to be those kind of power users. And then our persona is offering, what it's doing is it's taking all of those attributes of all of your customers and it's showing how are they naturally clustering together. So to your point about average order value, we want to be able to know is the, you know, urban mom.
Brent Peterson (07:00.654)
such as myself and my high value and also, you know, what are those other characteristics of me? Or is it maybe like a suburban, you know, early tech adopter, you know, so understanding those different segments and the value of them. So it's very important to know how to speak to these people. So we also want to let them know what products do they have affinities for. So, for instance, the products that I might be buying from the surf shop might be completely different from what you're buying.
So yes, it's important to know the value, but I would say focusing on that top 10 % will really drive your business in a meaningful way. So one mistake I see some marketers make is they spend too much of their budget just on acquisition and trying to acquire new customers. And that's like the saying, your most loyal customer is that that you already have. That's the easiest customer that you're gonna be able to reengage. So I think you hear a lot about.
that it just chatter from marketers today too is really leaning more into retention and finding ways to extract more value out of their current customers. One of the things that was a buzzword last year was client telling. So at the POS, you're also trying to get those customers that may have bought online, but they're just or they may not have bought online. Are you encouraging people that have retail to get those customers into their system and then into your system?
Yeah, we want to get all the data into the system. So whether that's that point of sale data, whether that's online data. So any of the, all we need is some sort of identifier. So again, like email, phone number, home address, but you want to resolve all those to those, you know, do that identity resolution process to figure out who these people are. But to your point, and I think you see this a lot more, even with D to C brands, they've realized that their most profitable customers are often those who want to go in store and purchase frequently. So you see instead of,
opening up a store in a mall in every town, you see them opening up strategic storefronts in those key markets. But it's very important within your data platform to be able to understand is this someone who prefers to buy online or is this someone who prefers to go in store because then you can present them with the offer that makes the most sense. In those customers, are there segments that you're still trying to solve? Is there...
Brent Peterson (09:17.166)
Is there specific things that are coming up year over year that everybody's trying to solve and then you're circling back and having to redo that? I think what we found is the most impactful thing is to really help use predictive analytics to find not only who are your most valuable customers today, but who are those people that have the highest propensity to purchase in the next four weeks, eight weeks, 12 weeks, or who are people that are predicted to be.
high lifetime value. So to be able to understand that instantly, so as soon as you acquire a new customer, to be able to map them to the right bucket and the right persona and to understand how valuable they are, it makes a marketer so much smarter. Similarly, when you think about the ability to activate on this data and really make an impact, you want to make sure that the data you're sending into those acquisition platforms, your Meta, Google, et cetera, it's the best data possible. So it's like,
it's the old adage and I'm going to screw up the language. But it's like, if you put good data in, you'll get good outcomes on the other side. So if we can shorten the timeframe that it takes for those acquisition platforms to read that signal and find the best people for your brand, then you're going to save money. You're going to reduce marketing waste and you're going to find more of those high values. So I guess the biggest change that I'm seeing, what I always encourage people is again, lean into finding.
those those high value folks understand who they are, what makes them different? What are their different affinities? Like what products do they want to buy? And in what order? And what's the life cycle? So how many days before they want to see the next message from you? But yes, really kind of leaning into the predictive capabilities is big. One of the um,
One of the mistakes I've seen merchants make is in that predictions that they get bad data and then they make predictions on bad data that just gives them bad outcomes. How do you ensure that your, your, maybe your cycle has to be fast enough to know that that outcome is bad or the, the, the test is the hypothesis is right. Yeah. I mean, that's super important. We're actually just releasing our.
Brent Peterson (11:24.174)
next version of our predicted models that are launching actually tomorrow in our platform. And I think one of the practices that we have is we're always checking retroactively how accurate were those models and also doing it at kind of that individual customer level and then pulling that all together versus just an aggregate model across everyone. So you have to make sure that your models are accurate, that you are checking and continuing to refine them and that it's based on those individual purchase patterns. I think earlier you said you
supplemented data to help enhance it enhance your your your core data right other other marketing tools are doing the same or similar thing are you seeing an acceleration in how the machine learning is working when you add that more data in and you get better results yeah and I would actually say it's pretty unique in our space that people are doing that data enrichment
at the platform level. You might see a marketer purchasing some of that data themselves, but that could be very expensive. So getting that enrichment data, so you're appending those additional characteristics, as you said, like demographics, psychographic, behavioral insights, but also things like survey data. So one of the tactics a lot of our clients are using today is,
they have an MPS score, so kind of a net promoter score for their customers tied to each individual customer. So we can pull that in too, as an attribute to understand how does that, if someone has a positive affinity for your brand, does that mean they're high value? Not always, but it's another good data point to have. So the more you can do to provide the fullest view of those customers, the better you'll do with your marketing. There is a bleed into B2B now from D to C, from B to C.
Are you seeing B2B using some of these same tactics yet? Our focus is certainly with brands who are selling direct to consumers. Because again, everything is at the individual person level in our platform. So of course there are companies who are going after more of the business or householding. But we're pretty much leaning in on the individual known user.
Brent Peterson (13:38.062)
And so everything resolved that identity. OK. Going into 2024 now, and let's never use the word artificial intelligence, AI again. Actually, I like the word machine learning because that's been around forever, right? Right. It's not new. What do you think that the next big thing is going to be? Is there going to be a big thing this year that's going to help us kind of move up the notch? I think what I'm...
scene and what I feel is going to be kind of that next level. I do think that a continued focus on sustainability, whether it's related to packaging, shipping, product, I think consumers, especially the younger generations care deeply about that. So I think you'll continue to see marketers trying to find more sustainable practices. Another big trend that I think we're going to see more of is customized loyalty programs. So not just offering the same.
to every customer. So again, that's where personalization is so important. So if you know that one customer wants to use their points to get this product versus that product, you know, being able to have customized loyalty programs, I think you'll see a lot more of that. Obviously, and I know you don't want to talk too much about AI, but I do think that you'll see a lot more of the generative AI in the kind of the chat bots and the help. So when people for customer service and being able to kind of...
short circuit and make that a better experience for customers. I do think we'll start to see that in particular. And then more data and privacy regulation. So I always encourage everyone, make sure whoever you're working with to manage your data, that they do have the proper certification. SOC 2, Type 2 certification is very important. And I know that's a lot of mumbo jumbo jargon, but it is important to make sure that you're protecting that asset. Do you think the chat bots are going to be our new call tree?
And how do you get out of it? Like you can't just yell, representative, representative. As someone who often will need to call someone, I understand that. I mean, I think the hope is that maybe it's short circuits that like if they can understand those patterns, maybe it's a better experience, but it's too soon to say. Yeah. How much do you help your clients?
Brent Peterson (15:58.126)
I want to talk about chat box just for a bit because I've been on a couple of them where I just get stuck. Where they don't tell you what to do next and you're just sitting there and then you type something and they're like, then they go to the next whatever the next part of their thing is. From a customer experience standpoint, that makes you leave sometimes a brand, right? Like, I don't want to do this anymore, I'm going to go somewhere else.
Do you help them understand that from a marketing standpoint if that experience is bad? We're not directly involved with how they're executing those kind of chat box experiences on their website. We're more helping them to manage their first -party data, understand those preferences of their customers, help them build segments for activation purposes mostly. So we're less involved in the chat box, but I do think that everything's gonna need to connect within.
you know, a marketing tech stack eventually, and they're gonna wanna like continue to pull those insights in. Yeah, and I wasn't saying chat bots specifically, but I meant the whole experience that they're in that experience and sometimes that experience makes it worse than if they were to not have a chat. Yeah, when I heard a couple markers on some of the keynote panels today talking about just that one, thinking about the risks of AI, they said one,
risk is brand reputation. So if you don't generate a good experience, the way you typically would, that's going to kind of potentially tarnish the brand. So you need to be very careful about that. And then I think there were a couple other risks mentioned, but I do think that in particular that brand reputation piece is going to be something, it has to be executed flawlessly to your point or it's not going to be great. Yeah. My final question, going back to the question I asked originally about,
the impersonal AI, what I meant to say would be, what I meant to ask was over -personalized recommendations where it feels a little bit creepy. Yeah, well, and I think that's a little bit of what everyone's trying to solve for with cookies going away too, right? So you want to make sure that, you know, you're not necessarily combining data, your data is getting shared with third parties who you've never given that permission to. So I do think there's...
Brent Peterson (18:18.382)
heightened awareness around making sure that the customers don't feel like it's a creepy invasion. I think that what I would always say is you've got to have that equal value exchange. If you're providing, like I said, like the right product, the right promotion at the right time, when they want to hear from you, then I think you'll have very happy customers. If you're not and you're missing those points, then it's a different story.
Carrie, as we close out the podcast, I give everybody a chance to do a shameless plug about anything you'd like. What would you like to plug today? Well, obviously, Decile, because I'm the CEO. So please reach out to us if we can help you with your customer data. And also, I'll give a shout out. I'm squeezing another one in for my husband's architecture firm, L Studio. So if you're based in DC or New York, please reach out to L Studio. That's awesome. Yeah. If I'm on the East Coast and I need to get a house built, I know where to go now, right?
Perfect, exactly. Carrie Lawrence, thank you so much for being here. All right, thanks Brent.