Interview #6: David Roger

As discussed in the last weekly issue, getting data in retail is super challenging. With that in mind, I had the pleasure of chatting with David Roger from Hetal Retail, about what he is building and the data points you can extract from one video. I am super excited to share this conversation. Thank you so much, David, for taking the time!

This is the sixth conversation in an ongoing series with movers and shakers in the retail industry. Is there anyone you want to hear from? Please be sure to drop their names 👇

The following conversation has been lightly edited for context and clarity.

Noah Sobel-Pressman: David, thank you so much for joining us. For those who aren't aware, could you please tell us more about Hetal Retail and your background? 


David Roger: Before [Hetal], I started a business called Felix Gray. We pioneered the blue eye glasses space back in 2016. Ultimately, it grew to a larger eight-figure omnichannel digital wellness brand. Felix Gray really focuses on grouping the relationship with technology and creating a proxy to back it up. We tell those stories and sell those products beyond just D2C and into retail. We were fairly early on in the [omnichannel business]. The omnichannel strategy started in 2019. By the time we exited, we were nationwide in Target, CVS, Best Buy, and a variety of other retailers.

[The problems] became overwhelmingly apparent as we navigated that transition. We went from being in the online world where we know anything and everything at any given second. The data is there for the direct consumer. The trick is to cut through the noise and make sense of it all. It's like you're watching TV with the volume at 100 and that same TV is off, [so you can’t see] what's actually happening. The whole point behind Hetal is to say, how can we help brands easily, affordably, and gradually understand the shelf. From that understanding of the shelf, then work to take action on this [data] and improve their business.

For [Hetal], we are really focused on the category management and merchandising workflows. Essentially, [we are] tying sales data to shelf data to help better understand what's moving incremental sales. Then, brands can have more persuasive conversations with the buyer or have better internal decision-making. On the merchandising side, we help with in-store correction. For people who are not in retail, they do not realize that these retailers are just as much sellers as they are platforms, the same as Meta or Google. If your CAC on Meta is not good, you can't just expect Meta to fix it. You have to put resources, time, and energy into fixing it yourself. The same thing is true with retailers. They try [to help fix issues], but just executing the shelf is very difficult. The labor is difficult to make sure it's done directly. 
Brands have to deploy merchandising budget for correction. Hetal just makes [the merchandisers] far more affordable and streamlined because we can audit the store effectively first.

NSP: I would love to dive into that further, especially from a brand perspective. Why is it important for brands to understand their retail shelf, and what should they be understanding about that shelf?

DR: Let's say you're a digitally native brand.
You start in the online world, you get traction, and now, as you're growing your online [presence], you want to have multiple points of distribution. You want to [avoid] online acquisition costs sometimes. [That leads you to think] you can scale the brand in retail. However, when you launch your website, you don't just expect people to come to the website and just start buying products from day one. I think it's also the same thing if you launch at a [retail] store, you cannot just expect that your sales are going to be amazing from day one. [Success is] a combination of a lot of things, from marketing to driving in-store traffic, leveraging promotions, and sampling. It also has a lot to do with just making sure that the new store execution is correct and understanding based on what we're seeing on the shelf, this is what's moving up sales or moving down sales.

I have seen the mistakes of a lot of early brands, too. When they go into retail, they [think] this is just a switch that I get to turn on, and the sales just start to run. However, it is like any other part of your business. You have to really be managing it, putting in time, energy, and thought into actually growing that side of your business. If you don't know what's happening on the shelf, how can you really make good decisions.


But you walk into a store, and you are trying to buy a product on a shelf, and it's not available, then [the customer] just leaves, but no one realizes it. If you don't have that level of shopper visibility, how can you really understand if there is an issue or not. [Similarly] if I'm in a store and I'm selling for 20% off, but that signage isn't there, I have no idea if it's there or not. I don't know if I ran a promotion, whether the promotion was successful, or not, because execution is a factor. These things really matter.

The way that Hetal works is very different from the traditional merchandising system. 
Traditionally, you would work with the merchandising firm, and these are W2-approved employees who go into a store, see what's happening, audit the store, and correct the store all at the same time. Because they are going in, auditing, and correcting all as part of the same visit, it's really expensive. It's $20-$25 a visit. [Brands] have to visit 250- 300 doors in a given month. It is really expensive, and the data you get back is often in CSV and photo dump formats. It is really hard to actually make sense of what happened. 


Hetal flips that [model] entirely on its head. First, we take advantage of weekly shoppers who are already going to be in the store in a given part of the day or month. [The shopper] films the aisles on our behalf. Then, we scrape all the aisle information for all our different customers, and we package that into what we call an audit session. Each shopper is given a list of aisles we ask you to film. Each aisle you click into opens a camera natively in the app. It's a high-quality video, and all you have to do is walk down the aisle. Because the shopper is already gonna be in the store, we make it easy for you to find the product, and easy for you to survey the data. Shoppers don't need to fill out long survey questionnaires.

Our data cost is pretty affordable, but it's a video of a store capturing hundreds of customers or potential customers videos, which allows us to charge the customers based on pricing of essentially $2 per store per month. Instead of spending $20 per audit, you spend $2 in a month. Instead of getting a CSV or photo dump, that's where all the computer vision AI comes into play. Essentially, we take that video, we grab the right frame from the video to serve as a photo. Then, we braid everything in the photo, from inventory compliance to pricing compliance. We will analyze what shelf you are on and how many facings you have. We will show how placement on the shelf impacts sales. We will show geographic pricing. We are breaking down the data into actionable insights.

Then, for in-store correction, we have built out a whole different network of approved third-party merchandisers. Brands would historically work with them directly anyway, but now all they are doing is in-store correction. We will recommend a store list for their review, and then we'll push that work out to third-party approved merchandising vendors to actually do the corrective action. Now, brands are only visiting stores that are worthwhile, and are saving 70-90% for their budget. On top of that, because we're giving work to our merchandising partners every week, you have access to very fast lead times, [typically] two weeks or shorter. We will coordinate all of it, since it is time-consuming for brands to handle. Lastly, instead of getting that photo dump back to figure out what was actually happening, we'll grade the after photo that the merchandiser sends us when they correct the shelf. Brands can see what it looked like when Hetal was there and what it looked like when the merchandiser left the store. That is a great way of making sure that we understand how sales are being affected by execution, how to have better conversations with the buyer, how to have smarter internal decision-making, but also how to handle corrections at an in-store level.

NSP: As you mentioned, there is another part to this equation, and that is the retailer. I am curious what you think about how the role of the retailer in this merchandising process is changing over the next couple of years?

DR: Historically, and this is still very much true today, retailers are going to win more than the brand is. If I go into a store and one product's not available on the shelf, [there tends to be] another product on the shelf. I might buy the other product. If two brands are on promotion, but only one has its signage set up, I might buy the one that's 20% off [with signage]. The retailer still makes a sale, the brand loses. The retailer does not deal with the substitute problem in the same way that the brand does. Retailers have always relied on brands to help with the in-store execution.

Now, larger brands have larger budgets. The massive CPGs have the ability to at least deploy budgets to be really good partners in this way. [However,] it is still too expensive for them. They like to have better, more streamlined, affordable solutions, but they can at least handle this. Emerging brands, it's even more difficult. We always look at Hetal as a way of helping brands most thoughtfully design the shelf with their partners, like retailers. Then, when executing, we make sure that the plan is executed to the highest degree.

When a brand is in a buyer meeting, you can flip the script and pitch them [with the Hetal data]. The more data you can present back to the buyer to show that the ROI on different placements or facings, the more compelling that case becomes. Then, once you have won those things, you need to make sure that the execution is of a high degree. We know that the retailer can't be relied on exclusively to do that for you. Brands have to support that within store execution. [Additionally,] Retailers are pretty excited about this opportunity because we're building a brand-first approach. Then, at the end of the day, everyone is winning more with the solution. If the brand wins, the buyer wins because their sales go up there as well.

NSP: You have worked with many emerging brands. There are many emerging brands that subscribe to this newsletter. Launching in retail is a key part of an emerging brand's lifecycle. Is there anything you have seen emerging brands do that really helps them stand out and succeed in that account?

DR: I think it is not just winning the account. It is how you continue to win in the account that you've already sold into.
If you are winning with other retailers, that becomes really easy to then go to a different retailer and say here's a playbook that we've done and here's how we're seeing success. One brand [we work with] tied facing data to sales data, and said, when we have stores where we have two facings or more, historically, we're seeing 50% better on shelf availability rates. They were selling too quickly. They need to have more opportunities to have safety stock. They are able to successfully argue for a brand blocking opportunity. 


We have other brands that tied shelf data to sales data. When you tie shelf data to sales data, you can show how the shelf you are on impacts sales. They said when we're on a bottom shelf, we sell 10 units, but when we're on a middle shelf, we sell 18 units per week. Think what the revenue you could be making if we were on that middle shelf instead. Now, they are winning middle shelf placement for some of the best performing SKUs by doing that type of analysis.

That is a great way to win with a buyer. But also, at the end of the day, this is very, very relationship-driven. Buyers are busy. They have a ton of customers, a ton of brands that they're working with. They want brands that are basically helping their lives be easier by presenting them with what they think is going to push sales up. How are you helping the retailer with operations to make sure that in-store execution is at the highest level it can be. 


That is where Hetal, again, we're identifying these issues, and then we're saying only focus on these stores based on what we've audited. Historically, you're just relying on proxy data, like sales, and looking at sales that haven't happened for 3 weeks. Now you know there's a problem, but you've missed out on 3 weeks of sales. Having eyes on the shelf, having what we're building, a Google Analytics for the shelf, makes for far better decisions in terms of understanding what drives revenue and how to optimize that in-store execution. 


NSP: To take a step back, I would love to talk about the retail industry as a whole and where you see the entire industry headed over the next couple of years? Where do you see AI impacting the retail world most? 


DR: There is still $500B of in-store [grocery] sales that happen per year. Online will continue to take some aspects of that. But at the end of the day, fundamentally, margins are really hard to make online work in a lot of different industries. For instance, if you have to ship out cold product, keep it refrigerated. That means that there has to be an additional cost to the customer. In a way, it's just a lot cheaper to buy it, put it in your car, and drive home. Customers care about discovery, too. They want to be able to find out about this new flavor or try a sample in a store. There are plenty of grocery and natural grocery stores that do that well. Those experiences matter from both a costing standpoint, but then also from a general how do you let customers discover products.

From an AI standpoint, I think what we're doing is one of the many ways that retail can be improved. Retail practices are historically very pen and paper. There's a lot of manual work associated with it, whether it's trade promotion, deductions, cash management, or just the fact that this is the physical world. With new abilities on the computer vision AI front, we're able to turn a lot of the offline world, that physical world, online in a really meaningful way for the first time ever. That is really what Hetal is leading from a retail standpoint.

NSP: David, this has been an awesome conversation. Thank you so much for taking the time. The last question I have for you would be, is retail dead or alive? Why?

DR: As I said, I think that retail is 100% alive. From a margin structure standpoint, it's really hard for many brands to support the cost of a price point as competitive as a price point to the customer, if they are providing the landed shipping costs all the way to that customer. If they're tacking on shipping costs, then the customer's paying extra. The customer is happy to buy all their stuff in a store and then, drive it home.

I split time between New York and Utah. New York, it's a lot harder to do grocery shopping than Utah, but in Utah, you fit everything in your car, you put everything home, and you're done. You don't have to order stuff from a bunch of different places, get everything on Amazon, pay the additional cost, and deal with the packages. If something actually doesn't look as good, or you got the wrong order, you do not have to return it. There are pain points associated [with eCommerce], which I think that retail inherently helps overcome. Discovery is really important. You walk into a store, and you're like, oh, that's cool. I have made discoveries that I would not have been able to make if I were just perusing an online store. Online is always very search and intent-driven. I think that matters. 


The way that retail dies is if we don't keep up with making sure that customers are getting what they want. [For] customers, that experience is often very reliant on the technology that is helping support it. If you can't go on a brand's website, search in store, and understand those types of things, you'd better provide value in a variety of other ways in order to be useful. The same thing is true with in-store execution or making sure the promotions are set up, whatever that might be. The online world is always going to likely exceed and be ahead of the offline world, but we don't want that gap to be a 30-year difference. We want that gap to be a couple of years’ difference. I think that provides a really good experience for the customer. 
If it applies to a really good experience with the customer, then retail is not gonna die. The only way it's gonna die is if we shoot ourselves in the foot.

Was this forwarded to you? Sign up here.

Have an idea or want to chat? Respond to this email.

Is the email not reaching your inbox? Try this trick.