“Creating Artifacts” — with Justin Stewart, Co-Founder & CEO of Cherry Pick

Today, we’re speaking with Justin Stewart, Co-founder and CEO of Cherry Pick. We discuss Justin’s passion for data science, his time as a model, and how Cherry Pick is using AI to power the future of product development.

Today’s episode of Journey Map is brought to you by MindManager.

Journey Map listeners know that with the right map you can take your career anywhere you set your mind to. MindManager work management software puts your ideas, plans and projects on the path to success by transforming them into dynamic digital maps, charts, diagrams and more. Mind Managers flexible visual format makes it easy to capture, organize, understand and evolve critical business information and with powerful collaboration sharing features it’s the perfect tool for keeping remote teams aligned, on track and headed in the right direction.You can take a free thirty day no commitment trial today by visiting www.mindmanager.com/journeymap.

Welcome to Journey Map, the audio experience that deconstructs the career paths taken by some of the world’s most interesting people. Today’s guest graduated with a Bachelor of Arts in Mathematics and Economics from Reed College in 2012. He then got his Masters in Economics from the University of Pittsburgh. During this time he worked as a Research Intern at Lucky Sort and a Data Consultant at VirtueFi. He started PhD studies at the University of Pittsburgh but left in 2014 to move to New York to work in data science. He started as a Technology Consultant at Swanee Hunt Alternatives there he developed data science solutions in support of counter human trafficking efforts. He moved on to become a Data Scientist at Metis before leaving to found Adsthetic an influencer search algorithm and recommendation engine for brands and agencies driving improved campaign performance. Oh and did we also mention he was a fashion model. In January of 2019 he officially launched Cherry Pick the company to which he is presently Co-Founder and CEO. For those of you that don’t know Cherry Pick uses AI to power the future of product development. By applying proprietary AI to social images and consumer commentary, Cherry Pick enables vendors to measure the demand for products, before they launch. Essentially enabling sellers to get a glimpse into the future. Right now they operate exclusively in the beauty industry but have plans for expansion. Our guest is taking the beauty world by storm he is Justin Stewart Co-Founder and CEO of Cherry Pick.

Today’s episode of Journey Map is brought to you by MindManager. Journey Map listeners know that with the right map you can take your career anywhere you set your mind to. MindManager work management software puts your ideas,plans and projects on the path to success by transforming them into dynamic digital maps, charts, diagrams and more. Mind Managers flexible visual format makes it easy to capture, organize, understand and evolve critical business information and with powerful collaboration sharing features it’s the perfect tool for keeping remote teams aligned, on track and headed in the right direction.You can take a free thirty day no commitment trial today by visiting www.mindmanager.com/journeymap.

DH: Justin welcome to Journey Map and thanks for joining me today.

JS: Dave that was really quite the pleasure. It’s quite the pleasure to be on and that was just a great intro. I appreciate that, It’s probably the most extensive intro I’ve had so far in my life so I appreciate it.

DH: Well you’re pretty young still so in another few decades it’ll be twice as long as what I just said.

JS: Ha ha yep.

DH: Listen I want to start in the land of data science and economics. What is the first instance you had whether it was in high school or in college or even before high school where you realize that you’re really into data and in economics in general?

JS: That’s a good question so I would say that almost, like a prerequisite to an interest in data is an interest in behavior and people. I think the through line for all of the work that I’ve done in my career to date, all the studies is really just trying to understand why people do the things they do and what information or frameworks can you put in place to effectively forecast or understand what someone will do in the future or in effect understand what someone’s done in the past and why. I think that’s just my nature as a human being, I mean I’m the type of person who sits at a coffee shop and just watches as people walk down the street and if you see someone running you’re trying to understand like are they running from someone? or they try to catch a bus? you know are they late for a date? That’s just how my brain works and I think that my interest in data science specifically just was a consequence of investigating how does one acquire information to answer those questions that I want to answer? I think with techniques in the early days just the web scraping and just seeing the power available to essentially anyone with this tool kit was incredibly empowering. I could start to really dig into a bunch of questions that were important to me and matter to me with relative autonomy. Like if you wanted to do astrophysics or something maybe you need a laboratory, or you need a telescope, you need a bunch of materials. With data science it’s fairly simple it’s a fairly simple tool kit, and you can do incredibly powerful things. I think that really inspired me and just have been kind of on that path and trajectory since I guess yeah probably midway through undergrad.

DH: Your side hustle or side career as a fashion model I want to see if what the connections are other than the fact that now Cherry Pick kind of works in the beauty space when did that start and how did you get into it?

JS: Well I always like to make the joke because I think it is kind of rare to be able to make this joke that there’s this models and models phenomenon in my life, and actually just trying to make the case that they aren’t so disparate. Indeed being a model is something like trying to identify the archetypal perspective either on a garment or in terms of what’s the underlying data generating process that creates x phenomenon. I think there’s actually a strong through line there. I started modeling when I was a kid. I was probably thirteen, I was thirteen making more money than both my parents just as a model which was pretty crazy and empowering as a young person. I you know culminated doing things like I was in vogue, and walked in Paris fashion week, and kind of did all those things. Fundamentally I’ve always recognized that it was not something like I would ever dedicate my life towards, but more like something I had the privilege of having access to participating in and could create some you know wins in terms of providing access to things that other people don’t generally have access to and it you know was just fun. So I did that for a while but ultimately always recognize you know I’m a technical person, I’m of a technical mind and always realized I probably would want to go and take studying seriously. But, for a time period I wanted to kind of explore that and see where it got me.

DH: Well I don’t know any models personally and I only know of very few PhD’s or at least PhD candidates personally and now I know someone who was doing both at the same time so I’m a little intimidated but will carry on with the interview. So what was kind of the tipping point, like you mentioned that you kind of acknowledge it was fine but it wouldn’t necessarily be like kind of this lifelong career? I’m assuming like that level of access that you got to choose brands, to people, to influencers, to capital. You probably have done a lot of different things so why did you choose to go in the path that you did? Like even economics and data science are pretty broad categories so to narrow it down to where you decided to invest your energy in life what’s the story there?

JS: Yeah I mean if you don’t know much about the modeling industry it’s a fickle industry and it’s based on a lot of hype. I’m not someone who’s very interested in following what’s on trend and I am someone who is like a slow thinker, I think slowly about the world and I try to uncover things. It’s like that world isn’t much well suited to that type of person it was really something that I could dip into and have fun at. Again it was really just kind of a an opportunity thing. The decision to kind of move from there into data science was really easy even though it sounds like a very kind of windy road it was a very straightforward one. I think there is this kind of notion of influence in those spaces that I think is very interesting to me and I think relates back to what I’m doing today around social influencers etc, and kind of being around people with lots of star power if you will. I think is really interesting because especially given that kind of backdrop of economics, game theory, and microeconomic theory which was what I was doing in graduate school you know provides like an interesting layer on top of how economies function. Even simple questions like how is price determined in an economic market place? or something to that effect when you layer in these non-immediately obvious things like social influence right? Like go to a club in New York City and you find that 80% of the people in there are not paying and are drinking for free and 20% of the people there are paying exorbitant amounts of money. It’s like looking at these interesting circumstances where influence plays this outsized role on how prices determined in market places and kind of how these informal markets take shape, and better understanding how influence is paid for. You know, things that affect were always very instinctive and I think probably you might call it social capital there is a weird niche field of economics that I was very interested in around social capital. I thought given my kind of backstory on the modeling side but also on the economic side I had an interesting viewpoint and perspective to answer some questions in that space and I was what I was ultimately working my dissertation around before I ended up deciding to leave and just work on even more directly as an entrepreneur. I’m not sure if that answers that question, I’m not sure if there is like a direct answer to that question exactly. But it was an interesting and very valuable part of my life that has paid dividends when I walk into a room it has different implications given the rigorous training that I had by going to castings, having to do performances, and being on stage, and thinking about the world in that way. That changes how you open the door to an investor conversation I would say. I think it also had an effect like for example recently our company and myself I was just on the cover of one of the biggest trade magazines in our space ‘WWD’ Women’s Wear Daily is one of the biggest most important publications in the beauty and fashion space and they put us on the cover. I think it largely had to do with the interesting backstory of this A. I. perspective on beauty and fashion and how that’s informed by this rich history and background coming from the space and I think that helped contribute to the reason we could be on that cover, in things that affect them so I it is the interesting backdrop. I think fundamentally it’s still all around the kind of people and influence and thinking about how to model behavior.

DH: Well I have this whole story with one of our partners in New York, one time and we’re actually working on this problem which the idea was like could you create a platform that would allow say like a yoga studio that isn’t sold out for the lunchtime class to just push all these influencers by basically be like “Hey you wanna come do free yoga in fifteen minutes now?”.It was based on like you need to make sure that it seems like all the classes were full and if those classes can also be filled with people who are like exceptionally attractive all better. I liked the idea, so anyway I definitely would love to have a whole other follow up non Journey Map discussion about social influence from that side of things. But tell us more, what’s the elevator pitch of Cherry Pick and where are you guys at right now in its evolution?

JS: So Cherry Pick if you don’t know much about it we’re a social analytics tool for product developers, like you mentioned at the outset we’re starting in the beauty industry but we have plans to expand beyond, and the technologies on industries but to get to the level of depth that we’ve achieved you really have to focus.The problem that we’re solving is incredibly high failure rate of products when they’re launched in market. I’m not sure how much you know about the consumer packaged goods space but incredibly high numbers of products fail on delivery, it’s close to 80% of products fail. That might mean a brand developed a product and placed it in a retailer and it was never re-ordered because they didn’t get enough sales. That means product developers are consistently making products that no one wants. This is a really striking fact in the world that people take basically as just the way things are, and they’ve priced it in through promotions etc, etc. So what Cherry Pick is doing is working to solve the problem from the opposite perspective, can you actually evaluate how much demand is being expressed by a target audience or group before that product is launched to essentially guarantee that product will succeed on delivery. The way that we solved this problem is very sophisticated, what Cherry Pick does is we track individual products used in images and videos across social media so this is already an incredibly innovative thing. When we say products use that literally is not that this is a red lipstick in this post but rather that’s peachy pink lipstick by Too Faced cosmetics sku number 75 so we’re tracking the literal sku’s when they show up in images and videos. I can get into why that creates this kind of opportunity for a categorical difference in terms of the type of tools you can build once you have that level of granularity of data. But once we’ve tagged these images and videos with actual skus now when someone on social says something like “I need this, this is the best thing I’ve ever tried if I mix two pumps of that with you comes this I get the best moisturizer I’ve ever used before” you know we take all these individual demand signals where consumers aren’t literally telling us what they want but they’re reacting to the images and videos where these products are being contained. So when someone says “I would kill for that” we know they’re specifically talking about Fenty’s new lipstick. That connection between demand signal by consumer and actual products skus, today we track close to every sku currently being sold in beauty so we tracked about 75,000 products and when you include attributes it’s upwards of 450,000 to about 500,000 skus. We can actually effectively predict given a target audience how much demand they will show for skus once they’re released in the market. So today the way that Cherry Pick works is we partner with manufacturers on new brands from concept. So if Halle Berry or someone with a very large audience wants to launch a new brand what you can do is take her two million or two and a half million followers and evaluate the demand that they’re expressing for every beauty product on the market, and then reverse engineer down to the actual product attribute. What are the attributes that are going to be the most effective if she’s selling into that consumer group. Should she want a moisturizer, should she want a cleanser, should that be a cleanser that’s best position for Target or should it be a moisturizer that’s best position for Macy’s? We can actually answer all of those questions based on the attributes that we are actually seeing demand for across our target audience and move forward from there. So today we have this multi stage process that drives product development decision making from brand identity, to categories assortment, to ingredient attributes selection for those products all the way down to how to tailor fit a given brand’s assortment specifically to a given retailer that they’re positioning it to sell at. We have three brands coming to market this year so from a traction standpoint again we’re working with some top manufacturers globally utilizing our technology today to identify product risk and brand opportunities. We have three brands coming to market this year, our first brand I can’t tell you who it is yet but it’s incredibly exciting massive, massive talent. We’ve already evaluated this huge amounts of demand we think to be a huge opportunity it’s launching in two months we’re incredibly excited to see that first launch. Our second launch is coming in Q.3 and our third launch coming in Q.4. So it’s a really innovative play on this problem. I think most folks in our space have kind of tackled the problem with kind of typical techniques: you build an enterprise SaaS insights dashboard and you try to forecast what’s the next big ingredient for 2021 or something to that effect. You see a lot of tools to that nature in some form or another. What Cherry Pick is doing is very different we’re aligning ourselves with the incentives of the brand, we take a revenue share percentage off of all the sales these brands generate that are based on our data, and we actually integrate our technology incredibly deeply with the product development organization for the brand, and just drive a huge amount of the decision making that would otherwise be taken by a person. Just fundamentally we’re taking human process, human workflows and making them data driven in an incredibly powerful way. I’m happy to jump in and provide more details on any one of those aspects from business model or kind of the underlying technology that makes it all possible, but what Cherry Pick is doing is working to build the next generation of data driven beauty brands. Obviously then extend that into pets, then next into food and beverage, and then into alcohol, so we’ve got huge plans for kind of growth moving forward. Right now we’re starting in beauty, we’re super happy because there’s a lot of heat here it’s a growing market, there’s a lot of opportunity. There are lots of brands coming to market very quickly with a huge need for better insight and better information into what products will succeed and so we’re already getting some fantastic traction as we’ve talked about before.

DH: Well I don’t even know where to begin with all that because it’s all mind blowing in its coolness. There’s like six nuggets I just wrote down each one we could go into for ten minutes so I can say we need to have a follow up conversation at some point just to capture the depths of those things. As an example for me how do you go and negotiate that business model and take in a revenue share etc, but we could spend an entire episode just on that and because we have just a couple of minutes left, I’ll just say man I just need to keep following along the journey, it’s going to be awesome. If you could go back to that thirteen year old who signing their first modeling contract, getting woken up I’m assuming very early in the morning getting dragged to casting calls, failing a lot, getting a few breaks and you could tell that person hey fast forward to 2020 first of all you’re like locked down in quarantine as the world faces a global pandemic but that aside you can stay where you are right now this is where you’re going to get. Do you think that thirteen year old wants to know or do they want to explore life as it comes up?

JS: That’s a hard question, there’s a phrase that I’ve been thinking about which I think is maybe applicable here. I’m probably going to get a new tatt you know I got a couple of tatts, but only get this phrase “never look down” just don’t look down, just keep doing stuff don’t worry about it, if it works if it fails, all you gotta do is do stuff. Just create artifacts, create value, bring stuff to the table, move as quickly as you humanly possibly can, and just don’t look down, forget fear it literally doesn’t matter. A lot of people have kind of like choice paralysis literally it does not matter, go left or go right just make the decision. There was that one president or general I’m gonna say George Patton or something that said “A good plan violently executed today is better than a perfect plan executed next week” that is how you have to approach life just do things as fast as possible and create as much value as you can. That’s probably what I would say. I think that tripling down that create part I think like especially when you look at this culture of influence there’s this other side of it which I think is not so nice, there’s all of this kind of you know if you walk down Spring Street New York you know you see all these kids lives or identities that are kind of built off of the shopping bags they’re carrying and if they say supreme on them. I think that’s the negative side. I think like there’s a split there’s consumption and then creation, and it’s almost like you’re on one side of that or the other in different parts of your life for a different time in your life. I think what I’m trying to do and what I’d recommend to my thirteen year old self or just any human being in the world just try to stay on that creates side as much as humanly possible and it really doesn’t matter what else happens.

DH: Well first of all cool tattoo, second of all great message in general. I feel like I’ve probably said something similar to several people in the last few weeks actually as we’re dealing with this kind of global crisis we all find ourselves in. Our business right now for example is taking a real hard line on doubling down on growth at a time when the conversation of most small businesses is like cut, slash, protect, inforce all that kind of stuff. I think it’s coming from that same place of if you look at everything in life through one of those two lenses to realistically put it strikes we can either create value moving forward or we can just kind of like lock it down and consume whatever dread comes at us. Really insightful. That’s all the time we have for today. It was really awesome speaking with you, really cool to get in depth a little bit more than we normally do in terms of the actual business itself which I think it’s really awesome and people get a lot of value from it. If you had one parting message to share with our audience in terms of a piece of career advice you can give what it would be? and then we’ll literally end on that.

JS: Yeah I think I’d pick out one word from that last little rant that I just gave which would be ‘artifacts’. Think about the artifacts that you leave behind just if you did something, if you did a tutorial, turn that into some sort of artifact maybe it’s a blog post, maybe it’s a tweet, maybe it’s a model that you put on GitHub. Whatever you do just create as many artifacts as you can that are real and breathing that people can critique, criticize and get value from and see. So you can track that trajectory through time. So many people just kind of do things but don’t necessarily bring them to life, I think it’s really important to bring stuff to life and actually just kind of jump off the cliff and let it go.

DH: I think that Shopify’s Chief Operating Officer Harvey Finkelstein has this now famous quote which is like “Do things and tell people about it”.

JS: Right.

DH: Is like the equivalent of the artifact I suppose.

JS: Couldn’t agree more.

DH: Thank you so much I really appreciate it, let’s do this again sometime and all the best as you launch these brands this year.

JS: Thank you I very much appreciate it, it was wonderful to have the time and I appreciate the time.