Gathering data in the direct-to-consumer (DTC) space differs from gathering data through traditional retail channels. How do DTC companies benefit from their business model and how does it put them at a disadvantage? What information do they collect? In this episode of The Insights Factory, Ian sits down with Crystal Lee, Vice President of Data and Analytics at Wild Alaskan Company, to discuss in depth.
Gathering data in the DTC world differs from those selling through traditional retailers. How do DTC companies benefit from their business model and how does it put them at a disadvantage? What information do they collect? In this episode of The Insights Factory, Ian sits down with Crystal Lee, Vice President of Data and Analytics at The Wild Alaskan company, to discuss in depth. They discuss the various methods employed by the Wild Alaskan Company to gather data and how they link this data to their overall business KPIs.
---------
Key Quotes:
“We get involved and say, okay, do I think that what we're building and what we're saying we want to measure, number one, is that possible to measure? And secondly, will that give you an answer to say yes or no to your hypothesis?”
“The numbers are almost nothing without the story behind it.”
---------
Show Timestamps:
(03:17) Understanding environmental impact
(04:21) How do you know when data has made an impact?
(06:26) Gaining an understanding of the business question
(07:56) Thinking about data presentation
(09:24) Example of a successful data project
(13:15) Collaborating across the company
(14:38) Collecting and buying data in DTC
(21:38) Using data to inform the overall company strategy
(26:43) Presenting data to leadership daily
(29:52) Cooking show data
(35:48) Hiring for business understanding
(37:38) Telling a story with the numbers
--------
Sponsor:
This podcast is proudly sponsored by Seek, the leader in cloud-based creation and delivery of industry-focused insights.
--------
Links:
Connect with Crystal Lee on LinkedIn
Connect with Host, Ian Cook, on LinkedIn
Follow Seek on LinkedIn
Ian Cook: [00:00:00] Some days it feels like we're drowning in data, but data isn't knowledge. Data without context, or the expertise to understand it, is just something taking up space in your warehouse. That's where being insights driven comes in. Insights driven organizations focus on generating actionable insights, rather than just collecting and munging data.
Ian Cook: Insights drive better decisions. Welcome to the Insights Factor.
Ian Cook: Welcome to the podcast. I'm Ian Cook, the chief technology officer at Seek and your host today. We are lucky enough to have with us Crystal Lee, who's the VP of data and analytics at the Wild Alaskan Company. She's a strategic leader with over 10 years of experience in this field, working in e commerce, direct to consumer strategy, working for some huge brands like Walt Disney, Chanel, Kate Spade, Cole Haan, and now putting her talents to use at a direct to consumer [00:01:00] company called the Wild Alaskan Company.
Ian Cook: Crystal. Thank you so much for being here.
Crystal Lee: Thank you, Ian. Great to be here.
Ian Cook: Fantastic. Well, first let's just start off with, can you tell us a little bit about Wild Alaskan and your role as VP of data and analytics?
Crystal Lee: Yeah, absolutely. So as you mentioned, Wild Alaskan is a DTC seafood company, notably a mission driven.
Crystal Lee: company with the mission of helping humanity live in concert with nature and how that translates to business for us at Wild Alaskan is rebuilding the bounty that Alaska has to offer and making that more readily available to Americans and the rest of the U. S. A lot of people don't know, or you may, you may actually realize when you go to the grocery store that a lot of.
Crystal Lee: Your fish comes from other countries that's imported in, but Alaska has a ton of sustainable, delicious, wild caught seafood that is actually being [00:02:00] exported and not even being sold to Americans here at home. So we're trying to definitely make a change with that and starting with our DTC business, which is about five years in the making.
Crystal Lee: Has that
Ian Cook: traditionally been because it's hard to ship it down to the continental United States, or has there been other reasons why it's mostly exported out of the U. S.?
Crystal Lee: Yeah, it's actually, it's not that it's harder to ship, of course, because, in fact, we don't use any air shipping in order to be more carbon friendly.
Crystal Lee: We only use ground or boat shipping. But the real reason is because abroad. quality in seafood have a higher value to them. And so they're willing to pay a bit more for that seafood while the U. S., Americans in the U. S. are less culturally aware of this phenomenon. And so we're totally willing to buy the cheaper fish that came from another place that is not only less healthy, but less sustainable for sure.
Ian Cook: So has that been part of The day to day [00:03:00] for you is not just the data and analytics about how you're selling, say, through to, uh, whatever retailers sell, but understanding a lot more about your impact on the environment, understanding who is buying. That sounds like a, a much richer data problem than just figuring out.
Ian Cook: Like, Hey, we sold X hundred units in Northeastern Maine today.
Crystal Lee: Yeah, absolutely. And that definitely ties back to my work at Wild Alaskan. So my role being the head of data science and analytics and thinking through all of the, not only making the data available to everyone in the company, but thinking about how it makes a strategic impact on the company.
Crystal Lee: And of course, you know, thinking through. How we can sell to consumers, how we educate consumers, and how we get the product to their doorstep is all a part of trying to fulfill and further the mission that we have at the company.
Ian Cook: I really want to dig into something you said there, making [00:04:00] data not just accessible, but having it create a strategic impact.
Ian Cook: How do you know when data has made an impact like that, rather than just people looking at a couple of numbers and saying, ah, okay, now I know. To use my example, how many units I sold.
Crystal Lee: Right, exactly. I think that it's about having the right conversations with the right people at the right time. So the way that I see that our team, our data team makes the most impact is that we already have relationships with the different departments, department heads, and all of the people on those teams.
Crystal Lee: In that they understand the benefit that we can provide and come to us at the right time. And what we really tell people is, okay, when you're beginning a project or even thinking about planning for the next quarter of projects or the next year of projects, that's a great place for, um, the data and analytics team to come in and provide insights, right?
Crystal Lee: So we understand [00:05:00] The goal, the hypothesis of what you'd like to achieve, what your business targets are. And we're going to come back and say, okay, look, this is all the data you may or may not have known that we have. This is the insights that we have on our current member base, or we'll go out and do research on people that are not in our member base and say, okay.
Crystal Lee: Based on these insights that we have, uh, let's come together and think about what is going to have the most impact in terms of the tactics and the actions and the strategies that that team wants to take.
Ian Cook: That I think is core to so much of what I'm interested in, what we do here at Seek is this interplay between having someone come with kind of a business focus, business question, and then a data group saying, like, we know how to help you answer that.
Ian Cook: One without the other doesn't seem to be as. possible. So it's hard to call what you'd have as an insight. If you didn't have that, all right, what is the business question you're trying to answer? What is the, what are the open questions that you are looking for? Is that, do you [00:06:00] have a set process to go through that?
Ian Cook: Or is this just a, since you're a small kind of scrappy company, everybody's just talking together all the time.
Crystal Lee: We do have some processes for that in what we call an alignment doc. We generally follow very similar to agile methodologies. And that will come together with a document that covers what are the business goals?
Crystal Lee: What are the hypotheses we have here, or maybe what is the problem that we think we're trying to solve, what does success look like and what are those metrics for success, right? So, and that's again, where we get involved and say, okay. Do I think that what we're building and what we're saying we want to measure, number one, is that possible to measure?
Crystal Lee: And secondly, does that actually, will that give you an answer to say yes or no to your hypothesis? This is
Ian Cook: something near and dear to my heart. I come from a bit of an academic background way back in the day, and one of the things we talked about all the time was validity. [00:07:00] You can measure something, but is that measure or that metric you have truly measuring what you have a question about?
Ian Cook: And a lot of times the answer is no, like we were able to measure it because we could find data on it, but it's not truly getting at what we're interested in. It sounds like you're really working hard to answer that question on validity and making sure you're providing that kind of information. So let me ask this.
Ian Cook: When you say you provide insights. Have a presentation method that is mostly working with data and showing people maybe charts, or do they work routinely and come back and see like dashboards because they need the information repeatedly? Yeah.
Crystal Lee: So the end presentation or tool absolutely depends on the type of question or the type of need.
Crystal Lee: So when we do intake requests, we definitely think about what the output should be in this situation. Sometimes. Let's say it's the need to show people the proof that we want to go forward and try something new, right? And we have some [00:08:00] data to back up those assertions. Then we may create a presentation or in, you know, insert slides into a larger presentation that show that, well, sometimes we may do a postmortem also in a presentation that kind of shows the story of what happened and what were the results.
Crystal Lee: In other times, as you mentioned, they're ongoing and we need You know, a tool in our BI system for people to use. So for instance, one tool that we've built is for any promotion that we do, right? You can go in and analyze the profitability of that promotion later on using their LTV, using the cost of goods sold, or the amount of money off, things like that.
Crystal Lee: So that's more of a tool that you can play around with and teams can analyze on their own. How does this compare to our normal discount?
Ian Cook: Is there a particular instance or? Maybe one project that you can remember that where this played out particularly well.
Crystal Lee: Yeah, absolutely. So we recently [00:09:00] did some trials on the Facebook platforms and we were looking at several things.
Crystal Lee: One was. Do we do, so our normal discount was a 15 percent off your first order, uh, sorry, 15 off your first order. And then we compared that against, uh, giving out free product and different types of free product with different level of COGS. And then another one we did was a 25 off. And so we looked at that over time and we did find that.
Crystal Lee: Compared to the baseline 15 off product promos, we're giving us a higher LTV and a quicker time to pay back because, you know, the cogs for some, the value of something is a lot larger than just the cogs to us. Right. And that also helps them as. Since we're a membership, right? It gets them used to paying that same amount every month and not having one month lower and then the rest of your months higher.[00:10:00]
Crystal Lee: And then another thing we found was that in fact, the 25 off, which does extend the time to get to profitability, right? Since you've given them a higher dollar off discount. However, that did, uh, decrease our CAC. So much on the Facebook side that that was also a good option because we were converting more members.
Ian Cook: Just to, for the audience listening, we've said a couple of things that I want to just double back on one COGS cost of goods sold CAC consumer acquisition cost. So COGS, the cost of everything that goes into, I think you have a box of some kind that gets shipped out. So everything that goes
Crystal Lee: into the cost of that fish, the delivery, the fulfillment, the packing, et cetera.
Crystal Lee: And then there's
Ian Cook: customer acquisition costs, meaning what did it cost you To get that person as a customer. So, these are all interesting things. And you just mentioned a minute ago, the ability to adjust your COGS, which I think is interesting. So, in terms of collecting all this data, you also seem to have a lever where you can [00:11:00] say, we can increase or decrease the COGS for a particular package.
Ian Cook: So, you're moving things in, not just like a marketing perspective, where it's 15 off or 25 off, but you can also say, the cost of getting it to that person, is 15. is something we can experiment with, which I think is really interesting.
Crystal Lee: Absolutely. Yeah. So by playing with different skews of what might be something that's discounted or that's free, then we can say, okay, well, there's a balance between what's interesting to certain people, like scallops might be interesting to some and spot prawns might be interesting to others.
Crystal Lee: They also have different cogs. That also fluctuates year over year based on the natural catch that is made since we're a sustainable company, we're never going to, you know, harvest more than is available. So all of that can be played with to, to balance what might be the optimal gross margin that you can get.
Ian Cook: So as part of your job, you're talking about not just acquiring and using data on the customers, but everything that goes [00:12:00] into and through the system. And so you are. And correct me if I'm wrong, kind of a central store for all of this data for so much of the company, which means there's got to be a lot of people coming to you with those kinds of business questions.
Ian Cook: And again, getting back to this idea that you're not going to be an expert in, I don't know, maybe the, the catching of the actual product or the logistics to go from a three P well, third party logistics provider. So you've got to work really heavily with those other people to figure out exactly what those are when they come to you and say, I need to see the data on.
Ian Cook: COGS and what it costs to get to Southern Florida.
Crystal Lee: Exactly. And that's why I love to work with the different departments, have a relationship with them, understand. And then we always recommend, you know, don't just come to us when you say, okay, I would like to measure this, or I'd like to set up a test.
Crystal Lee: Actually, we would love to be there to hear the context of what the campaign is or what, you know, what the initiative is. The better we understand it, the better we can. Put [00:13:00] context around the numbers we're seeing and also understand, you know, is that data clean, right? The more we understand, the more we go, okay, well, that number doesn't really make sense to me.
Crystal Lee: Let me go back and make sure that this data pipeline is all good. So this is
Ian Cook: clearly where your expertise in cross functional teams comes in, because one of the things I hear repeatedly and I've lived through myself is the please bring the data people in sooner so we know how to help you answer particular questions.
Ian Cook: So I think that's a really interesting perspective. I do want to get back to something that I feel like I've skipped over a little bit. And I remember now at the beginning of our conversation, I even misstated talking about having retailers. You don't have retailers. You're direct to consumer. So I'm interested in understanding how you collect data or get more data about how you're selling.
Ian Cook: So can you draw for me a little bit of distinction? I'm used to more of the CPG world where you get point of sale data. I know this many boxes of cupcakes sold on a Wednesday when it was rainy.
Crystal Lee: Yeah, we're really lucky to be in [00:14:00] DTC and I think there's a lot of, a lot more control that you have as a retailer, as a DTC and that's, you know, part, part of why I think this is an area that's growing in addition to DTC.
Crystal Lee: As well, you know, like being at the right place at the right time in terms of being a consumer package good, you know, important on one side, DTC is also important in that we have a lot of control over what we collect and also what users experience, right? So we own the website and we own the data that comes into the website and we are able to track a lot of things, right?
Crystal Lee: Within. The bounds of what is legal with PII, obviously, but a lot of behavioral things that we can track too that are anonymous that tells us how people are using the website. It tells us how many times people tend to visit a page. We can use That data, plus we have a direct line to the customer in that they have given us their account information, their email, [00:15:00] and we have the ability to talk to them directly rather than for instance, a CPG where let's say Walgreens has that information, but you may not have that information.
Crystal Lee: And it just, yeah, it gives us a lot more. opportunity to play with personalizing that experience.
Ian Cook: And it sounds like, Ben, you become a little bit more of the expert in exactly what it means for someone to be on your website. If I'm, like you said, scallops, and I'm really trying to decide between scallops or the spot prawn, you can see people making those substitution kind of questions.
Ian Cook: Do I really want the... The scallops this time, and we really want the prawns and figuring out, you know, where people make those decisions so that you, when you say work with the cost of goods sold, maybe, Oh, it seems like people spend a lot of time trying to figure out the two. Maybe if we lower the price of one, we can get them to buy both.
Ian Cook: Suddenly you have a net
Crystal Lee: gain. Exactly. And we can do a lot of testing and learning. And we have that instant feedback of, uh, what are, what are customers doing that provides us, uh, quick information to understand, you know, what, what's working for people and what's not. And [00:16:00] then. In addition to that, we also collect data that people, personal data that people volunteer to give us, right?
Crystal Lee: So we have, for instance, new member surveys where we're asking them, you know, how would you rate yourself and your ability to cook, uh, wild caught seafood or how many people live in your household, things like this. And we get a really surprisingly high response rate on those. And so it's great to have consumers tell us exactly where they're at.
Crystal Lee: These are not things that. You could even buy from a third party vendor for sick because nobody else is asking how good you are at cooking wild caught seafood. But it really helps us understand, you know, like where they're at in their journey, what content we can, can we provide them that helps them uniquely, right?
Crystal Lee: Someone who's a super beginner versus someone who considers themselves a professional chef. They need totally different recipes and content. And then we also see how that then, over time, affects their LTV, right? And tells us where we need to focus.
Ian Cook: So I can see this [00:17:00] being really interesting data that a lot of CPG companies would want that can't get it because they're working with retailers.
Ian Cook: What about the other direction? Is there any data out there that, you know, if you could just magically Grab it. You'd really want to see.
Crystal Lee: Yeah. There's still a lot of data that we don't generate internally. And that's a lot when we go out to look at. acquiring new audiences, right? So a lot of the data that we do have, right, we can get it so called for free by asking our members.
Crystal Lee: But if you're not our member, if you're a totally different segment of the market, and we'd like to understand you better, that's not something that we have because we haven't collected it personally. So that's definitely an area where we still rely on, uh, third parties to help fill in. You know, what are those personas?
Crystal Lee: Where are the other areas of the markets and how are they behaving?
Ian Cook: When you collect this data, it sounds to me like there's a real challenge in getting it all into a [00:18:00] single place. Do you, so you're acquiring a little bit of data, you're collecting a lot of data, but you're also collecting a lot of data from the people.
Ian Cook: that have other sections. So I'm not entirely sure how you grab data from the catch. Do people hand enter it or do you have like APIs into some form of boat counting fish AI camera?
Crystal Lee: So we actually don't extend currently that far down all the way to the boat. Although our CEO. He is a third generation fisher, comes from a fishing family.
Crystal Lee: So he totally understands all of those, but we do get industry reporting that tells us the total available catch. That is the amount that is allowable for the boats in that area of Alaska to catch that would leave enough to make sure it's sustainable for many, many generations to come. So we do look at that, consider that in terms of which partners we work with.
Crystal Lee: [00:19:00] We. We do currently own a processing plant, so that's something we're integrating in to be able to understand and have more control over a vertical, vertical integration. But certainly there's a lot of data that goes into it that's already very well managed by the state of Alaska that ensures the sustainability of the product that we're selling.
Ian Cook: So it becomes a data normalization and data connectivity problem. Getting it all in, making sure it works with each other, making sure it all resonates and that you can work with it in any one place. Absolutely. That your analysts have one place that they can see all of those things.
Crystal Lee: Yeah. And there are certainly challenges in a lot of those areas in operations that you'd be surprised how much it is still handwritten or just described to you.
Crystal Lee: APIs are Few and far between for those kind of sources. So a lot of times we still do a lot of transcribing and, you know, like looking for ways to automate as much as we can. But to a [00:20:00] certain extent, you know, like those operational industries are not, not looking to deal with APIs.
Ian Cook: So I can understand that a lot of this data goes into affecting day to day things.
Ian Cook: Should we increase the coupon? Should we, uh, run an experiment on a website or did this experiment work? Not. I'm thinking more of going up the chain of, um, strategy. How often and how much do you use data to inform the overall strategy of the company? And can you give me some examples of where that might show up?
Crystal Lee: Yeah, absolutely. So it goes all the way up to setting our company wide KPIs. So at the very top, you know, like, and this is very much a partnership with finance in being able to. Build from bottoms up, what is affecting our actuals and what's affecting our forecast, and how do we then build upon that to create the targets that we're looking so that we're growing year over year as a company.
Crystal Lee: Prior to having all of this data in an easy to use [00:21:00] place and very accessible for our teams, you know, there's a lot of, there's a lot of manual work or, you know, like Taking a percent of what we see and applying that further down the chain. But now what we can do is actually build everything from the bottom up.
Crystal Lee: Say this is exactly how many SKUs that went into how many orders that resulted in how much revenue, how much exact COGS was related to those exact orders. Every piece is building from the smallest building block. And we're building up to say, okay. This is a pretty accurate view of how we got to this number, and these are the levers that we can pull that will have, you know, whatever kind of impact.
Ian Cook: I know tons of companies, including ones I've worked at, that would kill for that kind of lineage to be able to just say, here's how it rolled up to every single level and got to affect the. Like you said, the key performance indicators, your KPIs, and you can measure that and say, this is exactly how we're doing and see the value of that data really, really [00:22:00] quickly.
Ian Cook: That's fantastic. Yeah.
Crystal Lee: Our auditors also really like it too.
Ian Cook: I don't doubt it. You've mentioned for Wild Alaska, environmental issues are key. You've said, described yourself as a mission driven company. Do you focus on data as well and sort of the larger environmental impact of fishing, your percentage of the fishing, the situation in
Crystal Lee: Alaska?
Crystal Lee: You know, I would love to do more and more of that. We do at some baseline. So we understand how we're shipping, the carbon impacts of those methods of shipping. And we've done a lot to make sure that we can access the entirety of the U. S. by ground shipping, ground two days. So it's one day. That is our ideal, but within a two day ground shipping radius, we can get shipments out to every single doorstep, uh, in the U.
Crystal Lee: S. And so that's been, um, a strategy from the very beginning, uh, to really look at where are our fulfillment centers? How does our [00:23:00] procurement and fulfillment network look? Um, the operation team works very hard, um, in working with last mile carriers, uh, so that, and the reason we do this is because. Ground shipping, not only does it cost more, but it's a whole lot more fuel usage, a whole lot more carbon.
Crystal Lee: And we want to avoid that at all costs. So you know, so that no part of our, uh, supply chain is relying on something that's so carbon heavy. And we'd love to continue to expand that towards other areas of the business.
Ian Cook: Is that something that can react quickly to the data? Can you see something creeping up and then, and then shift fairly quickly in terms of your logistics or your supply chain?
Ian Cook: I always think of supply chains as, you know, very hard problem to solve. I've worked in logistics in the past and not the, the fastest to adapt to changing situations. So it sounds like you're constantly evaluating this.
Crystal Lee: Yeah, absolutely. And we do this by, you know, we have our [00:24:00] overarching KPIs, but then we are always looking for what are the leading indicators that are going to help us see those red herrings of what's happening, right?
Crystal Lee: So we do have APIs into most of our last mile carrier. And so we can look at. For instance, uh, the issue rate that's reported back, the delay rate, um, of every single carrier. And we have our own contracts with every single, um, fulfillment center. Um, especially, this is especially important in, um, cold chain storage and cold, cold, uh, like, you know, frozen fulfillment, um, as all of our fish arrives frozen.
Crystal Lee: Um, that we're keeping a daily eye on those kind of... More leading indicators say, okay, well, if this is out of the norm compared to the last week or the last month's worth of our data, can we, and then have all the right dashboards in place to say, okay, and we saw that it came from. This fulfillment center with this fulfillment partner or this last mile carrier, then we immediately [00:25:00] go and talk to that partner, um, and, and think about what to do.
Crystal Lee: And if it's, you know, like it can escalate to a point where, okay, we need to switch last mile carriers, but we have a diversity of them in order to kind of spread the
Ian Cook: risk. Are there people in your company, you specifically and others who are constantly looking at those kinds of dashboards? Do you have sort of a, an ongoing control panel of this type of information?
Ian Cook: Is that, is it used across the company?
Crystal Lee: Absolutely. Yeah, it's the first thing we look at. So every morning I give a five to 10 minute spiel with our leadership team in our daily standup. And we look at all these business critical stats to say, is there an outlier? Is there an incident? And this covers. It covers operations, it covers finance, CPA, which is cost per acquisition, the same thing as CAC.
Crystal Lee: So we look at that, we'll look at our customer service center and how many calls we're bringing in.
Ian Cook: Without giving anyone any names or company sensitive information, what was the last outlier that you saw that everybody was like, Oh, [00:26:00] we've got to do something? I think
Crystal Lee: one recent interesting outlier is that we saw A huge usage of one promotion code and it was linked to a certain type of Revenue, like a type of action that members could take on the site.
Crystal Lee: So revenue types could be like gifting or new revenue or renew revenue, things like that. So we saw a spike in revenues. And so then we said, okay, well, what was the campaign that drove this? Was this a super success? Right. And as we dug into it, we realized, okay, so it turns out that it. was very successful in terms of visibility, and we saw it, we saw it play out in terms of the usage of redemptions and the number of people placing orders.
Crystal Lee: At the same time, it had some effects on inventory and supply chains. So then we were able to then quickly pivot, get a group of people together and say, Okay, in order to be able to service [00:27:00] So you're saying that in inventory for this item, like what are we, what are steps we're going to take today to make sure we can fulfill these on time to our customers.
Crystal Lee: Oh,
Ian Cook: fantastic. So another question on this kind of thing, what is the most surprising bestseller you have that you wouldn't have expected but sells really well?
Crystal Lee: Ooh. I mean, I don't know if it's surprising. Our scallops. do really well because they're delicious and they're giant and we only have them sometimes of the year.
Crystal Lee: But having tasted it, if you had tasted it, I don't know that you would call it surprising either because it's just an amazing product.
Ian Cook: I, I love scallops. I also happen to love cooking shows and I know that scallops are the one of the things that If you are a risk taker on Top Chef or some other show, you'll do scallops.
Ian Cook: 'cause the chance to completely mess those things up is so high. So I'm a little surprised about the scallops 'cause I would expect that there's a lot of [00:28:00] concern about trying to cook those at home. Like I, if I had these gorgeous, huge scallops, I like, I don't know if I can cook these the right way. So clearly people are willing to take that chance and, and enough of them are, are not overcooking their scallops.
Ian Cook: That's fantastic.
Crystal Lee: Yeah. And I have. Some really interesting data around this exact cooking shows, scallops, and cooking them at home, because we collect data on all this, obviously. So
do
Ian Cook: you know, like, when people are cooking scallops on, like, Top Chef, and you could expect to have, like, a surge in scallop orders?
Crystal Lee: Ooh, I wish we knew that. I wish we knew that, you know, I should just go and write a little script to ping me whenever it's on Top Chef. I
Ian Cook: would bet somewhere out there, somebody has a data set on
Crystal Lee: what... There's gotta be a website tracking
Ian Cook: this. What, what dishes were cooked at what episode of the Top Chef?
Ian Cook: And like, how did the person do? And then you could just look the next day. I know correlation is not [00:29:00] causation, but still. Lean heavily in that direction. If
Crystal Lee: they both go up, I'm happy,
Ian Cook: right? Exactly. But what is the data you have then?
Crystal Lee: Yeah, so we actually do do live cooking programs that we offer to our members to join.
Crystal Lee: And we take data on the people who join those. And for one particular campaign, uh, was it, it might've been prawns or ground sockeye, things that are not necessarily like, especially for something like ground sockeye, it's Uh, ground meat, but it's not like ground beef. People might not be as aware of how to use that.
Crystal Lee: We actually saw that for those people who signed up and also did attend the Zoom cooking show, we had like a 50 percent uptake on purchasing. That's special. And then even for people who signed up but didn't attend the show had some uplift compared to the people who did not, you know, sign up at all. So for sure, [00:30:00] you know, learning a lot.
Ian Cook: It's almost like your version of the in store demonstration. Like I'm trying to think of correlations between The DTC and the true retailer focused CPG world where, you know, things that you can't do that CPGs will often think about is like placement on a shelf. Can I get an end cap? Can I get something going to help increase the visibility?
Ian Cook: But you've got to focus much more on, like you said, you did experiments on Facebook. I'm sure there are things on, you know, all of the social media. Well, maybe not all of the social media, but a great portion of that. But this is terrific because that's an interesting way to. Get people to think about how to use it, which, you know, you walk into a grocery store and you've got a couple of recipes up, you've got people chopping and showing you exactly how to use it.
Ian Cook: So that's really interesting. And that's not, that's really hard data to collect. Like I can't think of how somebody was like, well, here's how many people stood around and watched me make mac and cheese with Gruyere the other day, but you can get how many people watched, how many people ordered then the next day.
Ian Cook: So that's really fun.
Crystal Lee: Exactly. And it's, [00:31:00] it's fun to design these things. And yeah, if we think about. The ways that we can get to a single source of truth on our customer data and then design everything we do to optimize that in a non creepy way. Then, we can get a lot of great insights. I think the corollary that I might think of in SOAR is back when I was in MBA, we tested out some software that was using eye tracking software, right?
Crystal Lee: So then you might count up how many eyes were looking and how many were paying attention, things like that. Instead, on the website, for instance, we use scrolling and mouse hovering to see where the attention is.
Ian Cook: I have done both of those kinds of studies. That's really interesting. You brought up your MBA from Wharton, which people tell me is a pretty good school.
Ian Cook: The ability to pick where you wanted to head would probably be pretty high. Did you start in data and know you wanted to stay there?
Crystal Lee: Actually, I started back right after undergrad. [00:32:00] Joining Disney, which I think is what really piqued my interest in kind of B2C and consumer. And I started out doing corporate strategy and it wasn't till after my MBA that I actually entered into the data world.
Crystal Lee: But there were a lot of You know, like a lot of transferable skills over together. And then I've also picked up a lot more technical skills as I've progressed in this half of my career. Was
Ian Cook: there a moment or a project or a particular job where you said, I'm going to go work a lot more with the data directly?
Crystal Lee: Absolutely. I think that when I was at Cole Haan, we started out and that's. You know, it, the DTC part of Kulhan covers multiple channels, right? We have outlet, we have retail, we have wholesale, and we have DTC site. So there was, you know, a breadth of that data sitting around. And, you know, we, when I started there, we were at a much More immature place when it came to data in terms of being able to automate it, to automate the cleaning [00:33:00] of it, to make it accessible to people, right?
Crystal Lee: So I think that's where, when I first started working on that, and that's where I first started thinking about single source of truth and customer databases. And just, it unlocks so many things once you have that, or if you can get that into the hands of the people, your coworkers. Basically, right in a, in a friendly playground sort of way for their analysis.
Crystal Lee: And yeah, that is absolutely what, what kind of drew me in to get more and more technical and understand more and more, do more, more of the data engineering on my own to really understand like. all the pieces of the pipeline that actually impact the final product. So
Ian Cook: you came into data with a pretty heavy business understanding, especially for a particular company, and now you're running teams.
Ian Cook: Do you hire for people who have that business understanding, or how do you, if not, how do you get them that kind of business clarity?
Crystal Lee: Absolutely. I think that depending on the level, for sure, [00:34:00] Having some sense of the importance of the business objectives and general strategic thinking is important. So a lot of times during our interview, we talk about a problem solving or presenting something, especially with a homework assignment to not only analyze the data, but also present it in a compelling way.
Crystal Lee: And that's also something we work on with my team as they're growing. in their career, right, is it's so important for data to not be relegated as number pullers, data pullers, but to be a strategic partner. And the more that you can be open to thinking about that, to thinking about your partner's problems and how It affects the business.
Crystal Lee: The more you can be that strategic partner. And in the end, I think everyone in data would find that to be more interesting, uh, as their day to day life than pulling data. Data
Ian Cook: visualization has been a real revelation to me as I come from a background where we did statistics and econometrics and machine [00:35:00] learning and just kind of spits out a number and be like, well, this is the number for you.
Ian Cook: This is the number you wanted. But the storytelling around it has to be so much stronger than it's something I came to much more slowly in my career than, than the coding and just working with numbers. Sounds like this is a big part of the kind of training you have for the people you work with.
Crystal Lee: Yes, absolutely.
Crystal Lee: I mean. The numbers are almost nothing without the story behind it, right? And that's something that I was lucky enough to learn early on in my career. And it's, I think, not a coincidence that Disney, a great storytelling company, required from, you know, corporate strategy that you are telling the right story with the data.
Crystal Lee: And that's how you get deals done. The only way to convince someone that this is the way we want to go rather than the other way, or the only way to even have a robust discussion is to understand the story that the data is telling us.
Ian Cook: So Crystal, I want to say thank you so much for your [00:36:00] time and this conversation.
Ian Cook: Before we head out, is there anything that you would like to tell those folks considering DTC or currently in DTC on their products to think about with their data and how to approach the whole problem of getting the data that they need? Yeah.
Crystal Lee: I think that working on a way to get to identifying your customers and making it very easy to link that to business KPIs such as LTV or CAC and being able to automate that or let people play with that.
Ian Cook: LTV being lifetime value. So the
Crystal Lee: value of a customer. Lifetime value. Actually. Yes. So how behaviors. Impact LTB and allowing teams to be able to play with that as a tool I think is something that always sparks interesting conversation and can really go from data insights to action.
Ian Cook: Fantastic. Alright. Is there, uh, anything that you would like to say on where people can find you?
Ian Cook: Are you active on social media anywhere? [00:37:00]
Crystal Lee: Yeah, absolutely. On Facebook and Instagram as at Wild Alaskan Company. And the same for our website, wild alaskan company.com.
Ian Cook: Terrific. Thank you so much, Crystal. It's been an absolute pleasure and I wish you all the best. And I'm going to look at the Wild Alaskan website because some of that, some of those scallops sound fantastic.
Crystal Lee: Absolutely. Thank you, Ian. And, uh, hope to see you in my database as a customer soon.
Ian Cook: Absolutely. Thank you.
Ian Cook: This podcast is proudly sponsored by SEEK, the leader in cloud based creation and delivery of industry focused insights. Thank you for listening. If you liked it, please feel free to rate and subscribe wherever you get your
podcasts.