#161. Mastering Market Insights: Leveraging Particl’s AI for Smart Product Strategies | Podcast

Vira: 0:41

Hello, hello everyone, and welcome back to Email Einstein. I’m your host, Vira Sadlak, and you’re listening to podcast by Flowium. And today we have a very special guest with us today, Ben Reynolds, the head of partnerships at Particle. For those of you guys who don’t know, with Particle’s cutting-edge tools, you can not only grasp marketing nuances, but also identify golden opportunities, make data-driven decisions, and basically bring your business to new heights. Today we’re going to talk more about all of these amazing things that you can do with Particle. But before we go there, Ben, it’s great to have you with us. Please say hi.

Ben: 1:22

Yes, thank you so much for having me. I’m really excited to speak with you today.

Vira: 1:27

Yeah, I have, like, a gazillion juicy questions to you, but before we go there, let’s play a little blitz Q&A game just, like, to get to know you better. Okay, cool. Let’s do this. East Coast or West Coast?

Ben: 1:41

West Coast.

Vira: 1:42

Okay. Shopping online or shopping offline?

Ben: 1:45

Online, 1000%.

Vira: 1:47

Yes, we might be biased. We both are from ecommerce. We might be biased, but I don’t remember the last thing I shopped physically in stores. Okay, last physical item you purchased online.

Ben: 2:00

Oh, I recently purchased a pair of shears to cut my dog’s hair.

Vira: 2:09

Okay, good one. Good one. Pineapple on pizza, yay or nay?

Ben: 2:14

I know I’m going to get a lot of hate for this, but yay.

Vira: 2:17

Yay. I’m the same. I’m the same. I told you these questions are going to be super random, but it’s so fun to get to know you better. Okay. What’s the one item you’ve purchased for under $100 that has become an absolute game changer in your everyday life?

Ben: 2:36

Oh, that’s a good question. Okay, I’m going to go with Built. I bought some of their Luxe shirts recently, and they’re by far the most comfortable shirts I’ve ever worn. So I’m going to go with that. And that was a recent one, so maybe a little bit of recency bias, but I love them.

Vira: 2:54

Okay, cool. Guys, all of the links, all of the things that we mentioned in this podcast, as always, will be in the description box. Not that we like affiliated or them, but we just want to bring all of the good stuff to you. Okay, cool. And the last question that we always ask all of our guests, if there was like a theme song playing every time you entered the room, what would it be?

Ben: 3:15

Wow, that’s a hard question.

Vira: 3:17

That’s a hard one.

Ben: 3:19

I’m gonna say the, and this is lame cuz I don’t listen to a lot of music. I’ll be honest, I listen to a lot of NPR. So I would say the 1A intro song would probably be it.

Vira: 3:29

Classic. Classic. Okay, cool. Let’s do this. Let’s talk about Particle then. For those of our listeners who don’t know what Particle is or like what you guys are doing, tell us more about it, what it is, What makes it special?

Ben: 3:45

For sure. Yeah, so Particle is an e-commerce database, and essentially what we do is we collect all the useful tidbits of information off of e-commerce websites. So think product descriptions, photos, titles, pricing information, even down to the inventory movement. So that’s down to the SKU level. And the use cases for this are product development, product research, and pricing optimization. The nice thing about it is, one, there’s a lot of brands out there who use market data, which is great. You should always use data to validate your decisions. But historically, it’s been limited to in-store sales, Amazon, Walmart, kind of like the big marketplaces, and not necessarily directly from ecommerce websites. And the nice thing about this too is instead of having to wait for reports, it actually updates daily so you can make decisions quicker. And benefit from this, I would say, especially right now, is what we’re seeing from our customers is they’re selling less on those big marketplaces like Amazon because their margin is so much lower and focusing a lot more on their D2C site. So as that trend continues, you’ll have, you know, it’ll be more and more important to collect data directly from D2C sites.

Vira: 4:58

Yeah, I know. It’s really cool that the difference between this kind of like older tools that rely more on like outdated information, if I can say so, And tools like yours, they rely more on the data that is constantly updated. And yeah, that’s pretty cool. And I mean, we’ve seen many startups, many new brands failing due to basically poor product market fit or poor product research. Actually, I think the number is as high as 35%. 5% of SMBs, like small to medium-sized businesses and startups, fail due to no market need. And your tool is basically famous for helping brands with product research. How can businesses ensure that their product, that the product that they are considering align with current marketing demands and trends and how your tool can help them with this?

Ben: 6:01

Yeah, so historically, And from what I’ve heard from a lot of the clients that we’ve spoken to, market research takes a lot of different forms. It could be going to your competitor’s website and looking at their bestseller page, which may or may not actually have their bestsellers on there because a lot of times they just want to get rid of inventory. I’ve heard of people going to dance competitions for dancewear, seeing what people are wearing at the gym, even going to like Nordstrom and seeing what’s on the rack and just counting items, right? So there’s a lot of ways that people do market research, a lot of experience goes into it, a lot of gut feeling goes into it.

Vira: 6:39

Mm-hmm.

Ben: 6:40

What we found, though, is the ability to actually see what is selling, you seeing the actual movements of inventory, can be extremely helpful in knowing what the market is actually responding to. So for example, if I’m an activewear company, being able to see my competitors’ new product launches, being alerted to their new product launches, and watching how those perform can give me an idea pretty quickly of whether or not there’s a product market fit. if you’re comparing those to more mature products that they have, you get an idea of what the typical sell-through rate looks like, what the units sold on a daily basis look like, and then comparing that to a new product launch can give you an idea of what the fit looks like and what the actual demand is for that type of product. So using data to make those decisions doesn’t take away all the other types of market research that people do. You know, I think those are still valuable to some extent, especially whenever it comes to the experience of people who are running these brands. The idea behind having data like this is really just validating those decisions. I mean, 90% plus of new products that are launched end up failing. So if you can use data to make better decisions there and avoid the costly mistake of launching a failed product, and then using that data to launch, you know, a new, a new top seller can make a huge impact to your bottom line as a business.

Vira: 7:55

Right. But here’s the tricky question. And that’s like, more from like a marketing side of things. A lot of businesses use like tools like yours, a lot of businesses use AI-powered product research tools. Do we like risk oversimplifying that like very complex market dynamic? Do we risk getting into the market where all of the brands come up with like similar products because they’ve been provided similar data by all of this AI-powered product research tools?

Ben: 8:32

You know, I don’t, I don’t think so. And reason being is because the vast majority of our customers, like 99% of the customers, are not simply looking at a product and saying, okay, this product is performing well, I’m going to make an exact copy of it. A brand typically has a certain identity that they want to stick with, a certain price point they want to stick with. And so what they’re doing is they’re taking this data on like what colors are selling best. If they’re launching into a new type of product, what sizes are selling best and to what extent on each size. And then they’re making it their own, right? It’s not necessarily that this pair of leggings is selling the way it is because of exactly how it’s built. It may be a mixture of the brand itself. You know, the, the, you know, people buy from a certain brand just because it’s that brand. Like, I think Lululemon is a great example, right? Like, they can launch something and typically it’s going to do okay just because they’re Lululemon. But this data should be used in conjunction with your brand’s identity and make it your own. And that’s what the vast majority of brands are actually doing with this data. So I don’t think data will lead to a world where we’re all launching the same type of product. It will just give you the information that you need to make your product, which is your own, you know, in such a way that the market will respond positively to it.

Vira: 10:22

Okay. Okay. Yeah. One of the functionalities that you guys have on your website is on your tool is also competitor analysis. How can, like, in what ways does tracking inventory, like quantities down to SKU level, offer a competitive advantage? You said that you can, like, analyze what other brands are selling, like in what quantities and stuff like that. Like, how Does it help you?

Ben: 10:50

Yeah, I would say there’s one way that comes to mind immediately is from a white space opportunity. So if I’m a brand who’s selling a t-shirt in black and navy, right, those are the 2 colors that I have. Being able to look at my competitors and seeing what other colors are doing well for them can give me the opportunity to launch a new variant that could end up producing quite a bit of revenue for myself. The other way of this too is I think of like a brand, like a lot of bag brands typically go into apparel. It’s kind of like a— I don’t know why, but it seems like a pretty common trajectory for those brands to go from bags to apparel. And if I’m a bag company, I’ve never done— and I’ve never done clothes before, I’m really shooting in the dark unless I have data to back up what types of products I want to start with and how I want to expand into the future. And the benefit of this too is like you think of Zara as a good example where Zara launches a particular dress and if it does well, they launch it in multiple colors. But it takes a few months for them to really figure out if this is something that they wanna do and if it’s performing well enough. And I’ve talked to brands where they actually look at Zara and they see a dress that expands into multiple colors and then they kind of make it their own and launch a similar dress. that could take months and you lose out on revenue opportunities, right? Like if I launched that dress 6 months after it launched and was doing really well for Zara, I’m missing out on a pretty large revenue opportunity. So having this real-time data can help me react quickly to what the market is demanding.

Vira: 12:28

Right, and a lot of your customers are actually a lot of your clients, I guess, they leverage your tool for their product launch. I especially like the Chubbies example. Can you like walk us through that case study really quick, just like to show how brands can like leverage the power of like analytics in that product development cycle?

Ben: 12:54

Yeah. So Chubbies, for those of you who aren’t familiar with them, I imagine most of you are, but they really came out as like focusing on shorts and focusing on swimsuits. And one thing that they really wanted to do was get into pants. They tried launching pants a couple of times, didn’t go how they had planned. And so when they started using Particle, they used Particle to figure out what colors and sizes were performing best in the market. So looking at different brands that they kind of wanted to emulate in terms of the pants styles and looking at those to see the distribution of sizes and seeing which colors are performing best. And then they also analyzed pricing and sell-through rates to determine where they should price these new pants. So once they use that data and they launched their pants, they did over $1.5 million in sales in just the first week, right? So it’s a very data-driven approach that led to a very successful product launch.

Vira: 13:53

Yeah, and since you touched on that, like, pricing aspect of things, pricing is definitely a critical aspect of any product launch. Like, what’s the process? How does AI or how does your tool specifically assist businesses in optimizing their pricing strategies?

Ben: 14:12

Yeah, great question. So think about it this way. If you are a clothing brand and you have, say, 3 or 4 competitors in your space that you consider to be direct competitors, you may watch their, their price points fluctuate on the website, but there’s a blind spot in that you don’t know what actually happens to demand with those price changes. It could be that they lower their price or they raise their price by$10. And, um, nothing happens to demand. They’re just making an extra $10 on each sale. It could also be that they raise their price by$10 and they lose a lot, uh, in demand and end up losing money on that price change. So the benefit here is being able to see the backend of understanding what happens to demand when price changes rather than just reacting to a price change. I’ve seen, um, spreadsheets that companies use to keep track of price. It’s a very manual process of going through and seeing what the prices are for the products of your competitors. The nice thing is we’re actually using AI to pull those price points and put them in a very digestible view for the customer to see those price changes, and then also see what happens to demand. So price is important too, because, like, there’s a macroeconomic effect on price right now. Inflation’s high and prices are high.

Vira: 15:29

Right.

Ben: 15:29

It may be that in 2 years’ time, that’s not the case and price is lower. But the fact is, like, people are willing to spend a particular amount of money depending on what their certain situation is or what the macroeconomic environment looks like. So having that type of information and getting daily updates on price changes and knowing what happens to demand can be extremely important to make sure that you’re staying competitive from a price standpoint.

Vira: 16:45

Cool. And what about the seasonality aspect? Do you guys, like, take and take that into account? And what advantages does your tool have, especially when it comes to understanding that like seasonality aspect of sales?

Ben: 17:02

Yeah, great question. We, we very much push our data for Black Friday, Cyber Monday. Obviously, that is the busiest time of year for brands, right? And there are a lot of different discount strategies when it comes to that, that, that, you know, that holiday, you might say. So you think about like, you may have certain products that are a certain percentage off, you may have the entire website a certain percentage off. And what we’ve seen with the data is that the approach of just having particular discounts for particular products rather than a site-wide discount actually works better. And if you want more detail on that, I probably couldn’t give it to you. That’s just what the data tells us. And we’ve seen our customers implement that and have greater success doing that. So it’s very interesting to see what discounting strategies work best, and using that data can have a huge impact on your business during Black Friday, Cyber Monday, just because You know, that’s whenever brands bring in the most revenue. Cool. Cool.

Vira: 18:02

And since we started talking about this, like, seasonality aspect, preparation for Black Friday, Cyber Monday, a lot of people are actually looking into what their competitors are doing and they are trying to anticipate and respond to competitors’ sort of like moves, especially during that holiday season. So how can businesses leverage historical data to anticipate and respond to competitors’ moves?

Ben: 18:30

Good question. So I would— I— oh, sorry, were you gonna say more, Vira?

Vira: 18:33

No, no, no, go ahead.

Ben: 18:34

Okay, cool. Sorry. Um, so I would say like the, the nice thing about this is, uh, and probably our biggest moat when it comes to competition is historical data. Um, many of the brands that we’re tracking, we have 2+ years of historical data. So I could go back and see what, what types of discounts they did the previous year versus the year before that. Is there a pattern? Have they done the same type of discounting over multiple years? And if so, then I can anticipate that the following year they will likely do a very similar type of discount. But not only that, we can also see how well those discounts performed, right? It may also be where, you know, maybe we’re a new brand and we’ve never done a Black Friday, Cyber Monday discounting strategy and we’re looking for different ideas. If I find a brand in the platform that has done 2 different types of strategies, I can actually see which one did better for them and then use that strategy for my first year. So especially for like a newer brand, that can be extremely valuable because if you have, you know, your first Black Friday, Cyber Monday is really positive, it can have a huge impact on your business going forward. So I would say like the historical data piece is huge just because you can see how different strategies change over time and which strategies are working better than others.

Vira: 19:51

Interesting. Yeah, there’s this one little case study on your website. It’s not necessarily Black Friday, Cyber Monday related, but it still was interesting. Like one, like gym brand, I forgot their name. They basically launched their product based on the assumption that if their competitors are like launching this product, there is a big opportunity for it. But they actually only sold out like 1/4 of their like original launch. They basically realized that competitors’ sell-through rates and inventory levels were performing poorly at the end. Tell me more about this like case study and what could they have done differently?

Ben: 20:40

Yeah, so this is Jim Reapers is the brand that you’re thinking about who are doing fantastic. They’re growing at a pretty amazing rate right now. But the specific example is essentially they saw 2 of their main competitors launch around the same time, the same type of product. And so they assumed that there was a big opportunity there. They purchased 20,000 units of inventory. And like you said, they only sold 5,000 or a fourth of the inventory that they had in stock. So they ended up with about$300,000 in wasted inventory, which can be very, very, very painful for a brand. The learning point here is had they been able to— had they had Particle at the time, they would have been able to go see a couple different things. One, that those products were not actually performing as well as they thought they were. And 2, that their competitors actually launched with way less inventory than they had assumed. So had they launched, you know, maybe 5,000 units or 10,000 units, they could have cut those losses by quite a bit. And that’s the power of data really, is just You know, you act on a gut feeling and it can have huge impacts on the bottom line. But had they had the data to back up that decision, one, they may have launched with less inventory, or two, they may have just not launched that product at all. Cool.

Vira: 21:56

Can you like briefly walk me through the process of like gathering the data and like using the data as a customer? What happens when I like come to companies like yours? What kind of information can I get? How fast and stuff like that? I’m just like trying to understand the process, I guess.

Ben: 22:16

Yeah, for sure. So it’s a pretty similar process for everyone. Typically, we meet with a company, figure out what their goals are in the short term and long term. We want to figure out first and foremost if they’re doing product development once a year, if they’re doing it constantly. It just kind of depends on the business. Typically, apparel companies are doing product development more often than others. And then second, we want to figure out who do they feel like their main competitors are. In addition to that, we’ll do research on our end to pinpoint some other companies they may want to keep an eye on. And then essentially what they’ll do is they’ll select the specific companies that they would like to track. So we load those into the database for their specific account. They’ll have access to all of the historical data that we have for those brands and will receive new data on a daily basis. If we can’t track inventory level on the brand, we still bring in review information, bring in pricing information, bring in all the product information, just everything but inventory. And then from there, they’ll typically create collections. So for example, if I’m looking to develop a new product, I have the ability in platform to actually search for a keyword, and it will look across the entire database. And that way, I can pinpoint who are the top players in that space. I can take those top players, put them in a collection, use filters, So that way, if I’m looking for just leggings, I can, I can take out all the fluff and just focus on leggings. And I can start whittling that down to find what are the best-selling products, colors, and sizes, and then use that data to either validate decisions I’ve made internally or use that data to develop products. But it’s updated daily and pretty easy to use platform.

Vira: 23:59

So, and like, what kind of data can I get? You’ve mentioned like the amount of like products sold, I guess. Like types of products sold and stuff like that, what kind of like data insights I can get from there?

Ben: 24:15

Yeah, so if I’m looking either at a specific company or across, I’ll be able to see first and foremost, I can filter by revenue, I can filter by units sold, by price point, so on and so forth. I can filter by date. So if I want to look at the past 30 days, or if I want to look at the past year, I can filter depending on that. I’ll be able to see the performance of SKUs. So if I’m looking at a list of their top products. I can click Explore on one, and I’ll be able to see how the different variants are performing, so how sizes and colors are performing. And then in addition to that, I can see color distribution. So I’ll see, you know, how many variants they have in black, how many variants they have in blue, so on and so forth. And then seeing where most of the revenue is coming from, because you may see, you know, one brand has most of their products in blue, but most of the revenue is coming from black, right? So those are really important data points as I make decisions on what types of variants I want to launch. And then the second piece of it is, price distribution, which is, you know, being able to see where most of the products are priced, like in terms of buckets, and where’s most of the revenue coming from. And then even on top of that, gender distribution, right? So women, men’s, kids, unisex, so on and so forth. So like, I think of that one specifically, we work with a shoe brand, and they’re interested in launching kid shoes. And they were looking at their direct competitors. And we found like, yeah, their competitors have kid shoes, but very little, like very basically a tiny amount of their revenue is actually coming from kids’ shoes. So that actually led them away from, from launching kids’ shoes and focusing on what they had already. So lots and lots of different data points. And in addition to that, like sell-through rate, rate sold per day, price points, discounts, so on and so forth.

Vira: 25:56

Cool. I never thought that you can actually get that information. I thought that it was like sensitive data that brands don’t necessarily share. It’s just like interesting, what, like, how do you even get that information? Like, is that your prediction or you like actually have the numbers?

Ben: 26:15

So we actually have the numbers and we get it off of the sitemap is the most common way that we do it. We have a bunch of different ways to do it, but the most common way is going to the sitemap or looking at the code of the site.

Vira: 26:25

Oh, I see.

Ben: 26:26

And more often than not, you’ll find inventory levels in there so you can see, you know, each variant. What is in stock? I can go back a day later, see what, what that’s changed. So if it’s gone from like 66 to 62, we can assume 4 units moved. We know the price point, multiply that by price point, and you get revenue. And typically we’ll see, you know, we have brands before they sign up, they like to do data checks. So we’ll actually like pull up their page in the app and we’ll benchmark what we have versus what they have. The most recent one we did, that was just yesterday, we were 98% accurate on, on unit volume.

Vira: 27:01

So, yeah. Awesome. Awesome. Yeah, no, it’s like super important, especially now, um, during the recession and with the rapid change of e-commerce, a lot of brands, they might even be like hesitant to make big moves without like betting, better understanding the competition and market trends. So that’s, that’s huge.

Ben: 27:24

Yeah. Materials are expensive right now. It’s tougher to raise money. So any data you can use, whether it’s Particle or other means, I would do it. It’s very important to make smart decisions right now when it comes to product.

Vira: 27:36

Absolutely. Absolutely. Tell us briefly about the ROI, like regular decision versus like informed data-based decision, I guess, and ROI of your product.

Ben: 27:50

I guess we can go back to Chubbies. So Chubbies doing $1.5 million in sales in a week, they paid for the full year’s access to the product times 5 to 10 just in that one week by using the data to launch those pants. So the ROI is huge. Typically, whenever it comes to data, like, you know, an NPD, you may pay $100 grand for a report and it may be, you know, 3 months late or 6 months late. But, you know, price point varies depending on how much data you actually want to track with Particle. But, you know, it may be anywhere between$20K and and $100K per year. And if you look at a$1.5 million product launch in the first week, ROI is pretty big.

Vira: 28:36

So awesome. Well, thank you so much. And if you would like summarize everything that we talked about briefly, can you again explain the significance of having like real-time market data and how it impacts decision-making for businesses that you guys work with?

Ben: 28:57

Yeah, I would say first and foremost, the market is constantly changing, especially in a space like apparel where trends are fleeting and they move very quickly. It’s important to have real-time data to make quick decisions and be agile as a brand. And I would say like right now with the economy, it’s important to make very, very smart data-driven decisions to make sure that you’re avoiding costly product, you know, failed product launches and building more into your— you’re building more hero products. So, but that’s the main thing I would say though, Vira, is just like, the market moves quick and having real-time data allows you to move quicker.

Vira: 29:38

That’s a good one. That’s a really good summary. Thank you so much, Ben. It was so fun having you here. If people want to learn more about what you guys do, where should they go?

Ben: 29:48

Yeah, I will actually, I’ll give you a link that you can include with this, but I’ll have a specific landing page for anyone who comes through here for actually a discount. But main thing, we’d love to meet with anybody, provide them with some data, even if it’s for free. So that way they can make, get or start quicker making smart data-driven decisions.

Vira: 30:09

Thank you so much, Ben. And as always, guys, all of the case studies, the Dream Reapers case study, the Travis case study, everything will be in the description box to this podcast. Ben, thank you so much for coming. It was so much fun having you here. I’m sure our listeners will have a gazillion questions to you, and I hope they will be able to make better decisions using tools like yours. Thank you for coming. Awesome.

Ben: 30:33

Thanks, Vira. Appreciate it.

Vira: 30:34

Guys, as always, if you like what we do, please leave us a review. And if you leave us a review and send us a screenshot of your review, just like email it to us, we’ll send you some nice Flowium branded surprise. So thank you so much. Come back next Tuesday. Next Tuesday, we’ll keep talking about all things AI in e-commerce, AI in email marketing. It’s going to be a good one. Thank you so much for listening, and we’ll see you soon. Bye, guys!