x

READY TO BECOME A MEMBER?

Stay up to date on the digital shelf.

x

THANK YOU!

We'll keep you up to date!

Interview

Interview: Optimizing the Shopper’s Visual Experience, with Jehan Hamedi, Founder and CEO of Vizit

Having the right images on the digital shelf wins search and drives conversion on the product page. Optimizing images for performance, at scale, is no longer an art, but now a science at the fingertips of digital marketers. Jehan Hamedi, founder and CEO of Visual Brand Performance platform provider Vizit, joined Peter and Rob to explain how brands are perfecting what your consumer sees.

TRANSCRIPT

Peter:

Welcome to unpacking the digital shelf where we explore brand manufacturing in the digital age. Hey everyone, Peter Crosby from the digital shelf Institute here. Now having the right images on the digital shelf, wind search and drives conversion on the product page. Optimizing images for performance at scale is no longer an art, but now a science at the fingertips of digital marketers, Jehan, Hamedi founder and CEO of visual brand performance platform provider visit joined Robin me to explain how brands are perfecting, what your consumers see. So Jehan, thank you so much for coming on the podcast. What you guys are up to at visit is really exciting.

Jehan:

Thank you. I'm I'm really excited to be here and, and talk to talk to you guys have been big fans of Salsify and follow your journey for a number of years now and, uh, just fortunate to be a part of the podcast.

Peter:

Oh no, our pleasure. And so, uh, your site describes visit visit, uh, by the way, to our listeners with a Z, uh, as the world's first visual brand performance platform, it sounds super important. So can you, can you break down what that means for our listeners? W w what, what were you, what, what are you trying to solve?

Jehan:

Sure. So all of us in the kind of omni-channel retail space are entering a new frontier now where consumers are scrolling through literally skyscrapers worth of image content on their devices. Every day, I read an article that said, I believe from Ogilvy's creative director, that the average person scrolls through the statue of Liberty in a day. And if you think about that, like there has just been a paradigm shift and it's, it's really impacted the way. Um, and it's impacted consumers expectations of the shopping experience, such that design and imagery has become a central tenant and, and kind of component, and ultimately a driver of brand success. So where were our kind of angle on this story is that, you know, we've, we've realized that in this new digital, visually centered environment, the rules have changed. And what I mean by that is image has now become both the gateway to consideration as well as the most powerful virtual Salesforce at driving conversion.

Jehan:

And so I believe, and, and where V like visits approach on, on this, this really fascinating problem in evolution to digital shelf 2.0, is that to when brands need to elevate their understanding from what, what you, might've kind of referred to as the voice of the customer, to now the view of the customer and understanding what consumers see in the most optimal way to present your products to them, but to do that, you need data and a lot of it. And so that data is what will give you insights and sort of a new perspective on, um, the visual competition, how to outsmart them, how to understand them, who are you competing against in that scroll in that search set? So visit visual brand performance is the technology it's giving you the tools to measure and understand the business outcomes that are driven by your visual branding and by visual branding. I mean, your designs, your image assets, your, your, um, your packaging itself, uh, with consumers. So that's what visit is and what we're about.

Rob:

So that, I mean, broadly speaking, the challenge that you're going after is a complicated one, and there's a lot of moving components there. If you're a brand manufacturer and you're selling on Amazon or Walmart or Kroger or target, or wherever, just to make this real, there's a lot of the pains that you don't have control over. And so if, if you're advising a brand manufacturer on how to improve the shopper experience through better visual performance, let's, let's get concrete. What exactly are you telling them to do? What data are you providing for guidance on how they can improve their, their visual shelf? Sure. So

Jehan:

When you nailed it, Rob, there are so many different elements of the retail experience that you can't control. And then there are many that you can, but we're driving a lot of success with our customers because the first and one of the most important things that they're focused on this year is trying to add science behind things that could not be previously measured. And that's imagery. Imagery is the last unoptimized aspect, um, of the PDP, because it hasn't been able to be measured, um, in any discreet and detailed way. So what our system is doing is it's kind of the difference between saying here are images that meet, you know, minimum retailer requirements that comply and images that are good and effective at driving conversion with a particular audience. So the insights that we're giving our, our customers is really focused on that carousel and helping to optimize both your kind of hero imagery, your supporting carousel imagery, your informational assets, your lifestyle content, any visual touch point.

Jehan:

That is, I kind of think of it like conditioning, that's conditioning that visitor that's come to your page. Each of those touch points has a material impact on the ability of that page to convert. And so what our customers are using us for is both from a kind of image selection standpoint from understanding their kind of visual competitive set. So what does best-in-class and vacuum cleaners look like and who hasn't, who do we need to beat? What scores do we need to outpace and outshine the competition. Those that's some of the data that comes through our system, where, you know, before the way that images were kind of selected and chosen is very subjective in the old world where you might come up with a few different ideas, and then you're going to kind of, you know, go shoot, shoot a whole kind of pallet worth of product and come up with what will ultimately be a hero image will ultimately become kind of lifestyle images. And then that's what you got for your PDP for the next 18 months.

Peter:

Yeah. More often than not, it was the side of a box product packaging from the physical shelf. So yeah.

Jehan:

Yeah, it's been, it's been incredible. And I think that the thing that I have so much fun with, and as, as, as a CEO, as a, as a founder, is I'm continually surprised by ways in which our customers are using the data. There are things that we never would have thought about, like visit is literally creating motorcycles today, visit it's creating a digital deli for Tyson. Um, our customers have embraced the fact that the digital consumer is so elusive, but so powerful. And there's so much data on them that if you could harness that data and integrate it all the way to the ER, all the way back to the creation of new pack and the new photography of that pack, you're going to pre optimize this, literally everything that you do while you're doing it. So it's been a really, really fun, uh, kind of journey where, you know, um, visual insights, uh, both from a diagnostic perspective and then using those diagnostics and those insights to drive selection and optimization, um, has been really something that our customers have been pushing us towards. Many of them we share with you guys,

Rob:

It's an age old question. It's some customers don't have any images that all of its, at least the e-commerce team doesn't have any images. And it's, it's a scraping and begging and borrowing and stealing just to get the six images to fill out the gallery. But then there's others that have really robust content engines and digital asset management systems that have terabytes and terabytes of images that are out there. And, and in those cases, the question is, you know, which six images and, and those are, I would imagine that that's more your sweet spot, where you guys are you're, you're talking about the folks that are already getting your chin over the bar on quantity. They're already getting their chin over the bar on, uh, doing a pretty decent job on, on the, the basics of the PDP. But then they want to say, what's the best, how I do better than that.

Rob:

So that's so interesting. Moving back to your last point. I really want to dig into that a little bit. Sure. If you know what images, when then you're saying that you actually informed the package design of a new product or a new refresh. So w why is that important? I mean, th and the reason I asked that is because, um, a bunch of folks that Coca-Cola's, and the Johnson and Johnson, they're not really using product photography that much on Amazon or whatever anymore, they're using CGI images, they're taking the model and they're, they're creating a digitally optimized version of it. And so he's like, what is it? What does it matter that the physical packaging is actually designed in a certain way to, to win?

Jehan:

So we're looking at the, like the, the visual representation of that pack, so that that pack is transcending both physical shelf and digital shelf, but I can give you an example. So, so our, the way that our customer is going to think about, about their packaging and, and, um, uh, sort of their hero images is, look, this is the first moment of truth. This image needs to have enough tractor beam, like power and an enticement, uh, that I'm going to draw somebody in when I'm competing for milliseconds, uh, for their attention. And I need to communicate my value proposition and exactly as much time. So what our customers are really kind of focused on is they're saying, look like when we, if we're creating a new product, new product development is a big vertical for us just as digital shelf is. And if I'm creating a new product or I'm creating a new sort of phase of my product, and I can give an example, um, the data that goes into that graphic, that ultimate kind of the face of your product, um, one of our customers put it in a really fun way.

Jehan:

I wrote it down. I said, that's awesome. He goes, well, we have to put our best face forward. How do we do that? Um, in a measurable way. So we found that you can actually leverage a data from a system like visits to understand how you can not only sort of help shape a pack design in its earliest phase, um, and kind of get those insights so that as soon as you take that thing to print, that becomes the new face of your product for the next couple of years, that becomes the new, um, you know, uh, image of your kind of rebranded product, um, that it's going to resonate with consumers. And we found that when you do that, you're gaining some really interesting, not only new insights, but efficiencies in the process back to my point earlier about the old way. So visual selection design kind of image effectiveness has sort of lived in two camps in the past.

Jehan:

One is kind of driven by that subjectivity and kind of gut feel and kind of making your best, trying to put your best foot forward with the information that you have at hand when the business decision needs to be made. The other camp is extraordinarily expensive. It's traditional market research, it's doing panels and surveys and focus groups that take weeks to months where you're looking at only a small subset of the ultimate kind of consideration set that some brilliant designers had conceived along the way. So if you can get insights on anything that you're creating 99% faster at 97%, you know, the cost and get a conversion advantage, then literally every visual touch point that you're going to project to the world can now be measured in pre optimized because it scales the insight scales. So for our customers, why it's so important is because they're finding that their customers are producing more and more content exactly.

Jehan:

To your point. Rob, some of these companies are just like content machines. They found ways to use CGI to create content they're, leveraging Influenster and UGC that they're, they're using crowd source 99 design, like kind of content creation platforms. Um, but the better image wins and what our system has been able to prove is if you have better scoring and better performing images on each visual touch point, then your competitive set, you're going to sell more product every time. And so what why we, uh, at visit got kind of really inspired by this is we, we focus very much on, we designed our system with images and photographs in mind that was sort of the problem that we were approaching, because that's what some of our early customers had, um, had told us. And we got, uh, part of our kind of journey is we were, uh, accepted into an incubator that was sponsored by PepsiCo and PepsiCo said, oh my gosh, like, you guys have to apply this technology to pack because the pack is going to be the thing that ultimately is kind of one of the key components, key pillars of an ultimate visual identity for an entire product.

Jehan:

And it's going to impact a category in our portfolio. We get that wrong and we get all of the subsequent, um, kind of activations around that asset wrong. We're going to be kind of swinging with a, with a handicap and, you know, I want to, um, kind of from, from, from their perspective, kind of just really reiterate that it's, it's been a really cool learning process for us as we've seen different applications unfold. Um, and the one that our customers have been incredibly excited about is, is obviously on the digital shelf and figuring out, you know, how can we get an ounce more advantage, um, in that, uh, digital purchase environment, when every single day my competitors could be changing their content, which will directly impact my ability to convert.

Peter:

See, and that's the future that I'm so excited about. Uh, but I know scares a lot of our, uh, you know, of the brands out there because that the need to be continuously adaptive, you know, it's that old challenge and opportunity thing, right? That, that therein lies your margin there in lies your share of, of shelf, uh, Sheriff's sale when you are able to keep up with, uh, with the merchandising changes that are offered whenever a retailer changes their, their attributes. And additionally, I would imagine whenever you're receiving data that tells you this works better than that, how more quickly can you put that learning into effect, um, across your digital shelf? And so th I think we share a philosophy in that helping, helping brands get better and better at that by helping to automate it, or, or drive workflow that will assign things that we'll do it like those, that kind of sort of automation of this continuous improvement, I think, is the real, um, not we have to crack together as an industry.

Jehan:

Yeah. I mean, Peter, one of the coolest things that, um, so one of our, one of our early partners who, um, she, uh, she built her career at Kraft is an incredibly impressive woman. She worked, she was a, an EVP at central garden and pet, and now she's running a CBD business, uh, called Caden wood, uh, which has been doing some incredible work, um, in, in that kind of burgeoning industry. And we w one of the initiatives that we worked on with her when she was at, um, central garden is, was very much focused on optimizing kind of the, uh, PDP for a product that was oriented towards cat owners. It was called comfort zone. And it's, if you look at it, it looks like one of those kinds of Glade plugins where you plug it in, and it admits these, these kinds of chemicals that calm your cat.

Jehan:

And so in her mind, they had a very clear competitor. Um, it's this company called field away, and they had a very clear target audience. Like you're not gonna buy that product if you don't own a cat. And through our work together and kind of running these diagnostics and these optimizations, it took their competitors six months to figure it out that she and her team had leveled up their product content. And they got a 31% increase in their conversion rate that persisted. And then over time, they started to get more clever about different changes and transformations and augmentations to that product. So imagery is also a very silent killer to your point, Peter, on, on kind of being adaptive and agile. Um, and it's something that neces, you know, that we, we, as, you know, folks coming from from a research background or a brand background, a lot of times you might think like let's set the identity and then roll with it, but we really need to invert our perspective to the demand side, which is the customer. And if we can understand the customer lens and what they want, and specifically what they are seeing, we can figure out more data-driven and intelligent ways to compete on that new dimension, which is a visual.

Peter:

Yeah. And for our, our grand listeners, we always have to translate customer to consumer because their customers are often the retailers. Right. So just want to make sure they're thinking in the right mindset, that what we're focused on here is, is who's the shopper, what do they need to, as you said, discover, and then convert, and how can images drive that journey? It's super interesting.

Rob:

I wonder if the Tropicana team had had this technology before that rebrand, whether this would have prevented them from spending tens of millions of dollars, basically updating how they, how their physical packaging was set up. Well, it's very famous.

Jehan:

It's a very famous case study. Um, you never know, you know, it's, you know, AI is extremely powerful and we found this particular application of AI is extraordinarily predictive. And, um, you guys are actually hearing it first. Um, we, we haven't released this research yet. Um, but if our system was in place five years ago, it would have predicted the commercial success of white claw relative to its competitive set, because their packaging was so powerful with a very broad set of their target consumer base symbol, female, and millennial consumers. So it gets really interesting when you can do, you know, visual targeting effectively and understand like, okay, here's a new pack. I'm not going to show this to any real consumers. I'm going to show this to any real humans, but I want to get a, get a sense. Can I get the pulse of how this would resonate with these particular groups of people? And I want to compare that insight to other folks that I know I'm going to be competing against either in a digital or physical environment or a digital marketing medium,

Peter:

Or they could have just come to south Boston and seen all the crumpled white clock hands that are literally from the stats sports bar. It's a walking hazard. DNR is clearly that works. I want to dig into something that you said earlier. Um, and I don't know if it's a case study you can talk about, but that you said Tyson is building a deli, a digital deli. Can you, can you talk about,

Jehan:

I can, yeah. We, um, uh, elements of that, I can, uh, so it's, we, we just made an announcement with our partners at Tyson, which is really fun. Um, so kind of walking back, uh, the, the use case. So over the last year, Tyson's, uh, e-commerce sales have, have boomed, um, as, as many have, uh, I think it was like an 89, 88% increase in their sales. And, you know, they're kind of, they're faced with this really interesting challenge of, you know, how, how do we adapt our, um, kind of shopping experience for some of our key products to, um, the digital environment. And Tyson is such a fascinating example because the category that we focused on had never been sold before online on a Walmart and it's raw meat, raw meat, traditionally, as an experience, you go into a store and you look at the, you know, the cuts and you, you look at, uh, you know, you're, you're, you're kind of, um, visually inspecting and you're educating yourself as you're an in store.

Jehan:

And so to adapt that and create a, a digital jail deli, a digital butcher shop, if you will, um, on Walmart is a really hard problem because you're trying to create a presentation where there is no visual standards of excellence that exists. Um, they have to create it, um, which was a really cool exercise to be a part of with them and the team there is so innovative because they were looking at everything they're looking at, you know, how you photograph the meat itself, how is the meat packaged on what kind of tabletop is the meat shown and at what zoom at what angles, or is there seasoning? What about the badging and kind of the, the quantity, how do you present all of that stuff visually? Uh, it's something they didn't have any data on. And it was a fun, a really fun opportunity to, to kind of work on because, because of the fact that people on e-commerce eat with their eyes, they drink from their desktops. And so whether it's raw meat or, uh, you know, a bottle of Pepsi or a, um, your favorite snack bar, um, those visuals have the role of a million salespeople. They have to sort of reach out and entice you and convince you of your value proposition and why you're better than everybody else in a millisecond. And for meat, that's very hard. So it was a, it was a fun, fun project to be a part of and, uh, gonna going to be making some more, uh, news about it soon.

Rob:

How, uh, cross-cultural does this work? And one of, one of the things that's so interesting about that Tyson example is the fact that it's something new that doesn't exist. So you haven't been able to train, um, the Mo the algorithm, the AI against 400 meat delis, or a thousand meat dailies that exist online. You're literally doing the first one. So you're using some type of trained intuition from other stuff on a new application and hoping that it translates that's, that's fascinating, first of all, in and of itself, but where it gets me thinking about is the different cultural expectations around design and around what's, what's attractive in a way that, you know, even, even around functions around around the world, um, you know, famously for anyone who's spent any time in China, big Chinese e-commerce sites are like an assault on your senses in all possible ways.

Rob:

They're just overwhelming, you know, they're like the opposite of the apple aesthetic. Um, and then you have the U S the, the trend has been to be, to be a little bit more simplified, uh, over, over time and Europe, it just depends on the country. And then, uh, what foods you're going to find attractive where, you know, like my dad's Cuban, my mom's a Yankee, um, the foods, their houses looked pretty different, and they're not like all immediately attractive to the other, to the other set. So, uh, I, I got to imagine that there's, that there's some challenge here, even within the United States of saying, oh, yeah, this is the right way to display a meat. You know what I mean? Like if it's dependent, it's got to depend on the audience. So like, how, how cross-cultural is this, how universal, how audience dependent all that to, you know, what I'm getting at here? So,

Jehan:

Yeah, no, I it's, it's a really awesome question because images are the language of the world, but they can be translated very differently, uh, depending on where you are. So from a, from our perspective, you know, our, our key invention, the key kind of, um, technology that powers the visit platform is what we call audience lenses. And that's an algorithm that is learning to simulate and the, the visual preferences of a very specific group of people and our customers define those by demographics, by kind of, or consumer cohorts by, um, you know, attitudinal and psychographic traits I can talk about, but market is a really big one. And so I can share actually a couple of examples. So for Tyson, uh, for this particular initiative that was focused on the U S um, which is a lot of times where we're most brands will kind of look to first implement our technology is to get an understanding on the, on the ground, in the U S team, um, answer a key business question, get a result, and then scale up.

Jehan:

And so for some of our larger customers, we are looking across the world. We've built, uh, AI systems that have successfully simulated focus groups in Beijing without talking to a single human in China. Uh, same thing in Singapore. Um, we've done, uh, for Harley Davidson, one of the coolest opportunities that we've had, which we can now talk about is I mentioned earlier in the session that, uh, they're integrating our technology, um, into the product development process, um, visit is creating motorcycles. So when you're designing a new motorcycle and you've got sort of the, the body and the, the wheels and the exhaust and, and finish, uh, different kind of color treatments that you could make, part of their process is an understanding how their product selections and ultimately their kind of product development is going to resonate in other markets. So we're looking at German consumers, looking at other folks all around the world, and we found that to, to answer your question just very directly, Rob, there is a big difference and the visual preferences of folks, culturally, there's a difference here in the Northeast to rural Wisconsin, uh, where a lot of my family is from, um, to the south.

Jehan:

You know, we, we kind of are, um, as, uh, you know, looking at all of us, kind of from a social science lens, you know, we're, we're, we're, um, products of our environment. And from understanding how to visually communicate with a particular group of people, you have to understand what types of content they're exposed to on the internet, all of that content, all of that imagery is public, and it's minimal. It's a big visual data problem. It's an totally, totally new set of data and insights that can be gleaned. So the key is understanding sort of that digital zip code, right? Understanding kind of what is relevant and what are these different groups of people choosing to expose themselves to on the internet. If you can figure that out, you can get a really direct line into what they're most motivated by and what that you have, or what presentation of your products that that can be made would be most motivating to them. So it's been fun. We've looked at probably a dozen different markets now, um, and the, the, the China one was, was quite a bit of fun. And, um, for some of our, our friends in the cosmetics industry, there are significant differences in between the us and France, us and Germany, um, to how they can visually present their products by retail as well. Retail shoppers are different as you'd imagine, um, people are different

Rob:

Panda. So this, this really is a crazy optimization problem. I'm just imagining like a three-dimensional cubic bread in my head. Yeah. I graffiti culture, psychic graphic, blah, blah, blah, blah, blah. So you really are actually going after like a universal good versus bad. You're really going after. All right. Well, pick your, pick your slices. What are you optimizing for? Who are you optimizing for? What's the context. All right. What are you starting with here?

Jehan:

Yeah. Know your audience. Good, good. And bad is all in the eye of the beholder. So you have to understand the end consumer. And it's so fascinating because, you know, brand management and, and kind of marketing as a discipline, um, for ever has been trying to kind of put yourself in the shoes of that 18 year old consumer of that 25 year old consumer, and think, I think they would like this, but ultimately we aren't that consumer, we aren't the end consumer, we're making our best understanding and kind of, sort of best guess at what they're going to react to, can't see through their eyes. So with a massive amount of data and some really cool technology, it is possible. It's not only possible it's pinpoint accurate, uh, to be able to understand their view of the world. So it's, it's been something that I I've been on our team has been incredibly passionate about is as you'd imagine there's a, data's in our DNA as a, as a culture

Peter:

Sensing data nerd. I have designed well. Yeah, I'm not a cool guy. Uh,

Jehan:

But, uh, no much of, uh, much of the core team at, at visit has been, uh, building prior to visit had been building insight platforms and designing research process for some of the biggest companies in the world. Let me ask you about

Peter:

That. Cause I was, I was, are you interested in the, in the span between a Harley and a Tyson, like, do you have to be very in order for you to enter into a new category of manufacturer? How much work does that, like, could you basically work for any customer right now or do, or do you have to sort of decide where you're going to play because you need the right data sets, et cetera?

Jehan:

Well, we have a pretty flexible approach, uh, by design. Um, but we've in the system is flexible in that we're, we are serving and working with, uh, several different categories of CPG consumables, how we're working with durables, you know, uh, appliances, uh, obviously auto. Um, but what the system needs like

Peter:

Any, have you been in home goods or furniture or any of that?

Jehan:

Yeah, we have, I believe there's one in home goods. Um, but we've had a lot of our bread and butter has been in CPG and in companies like Harley Davidson and others that have a very, um, substantial content apparatus and a very sophisticated, uh, consumer insights function, um, in practice. So I think, you know, where, um, where we, the only limitation to the technology is your ability or someone's ability to describe traits of your target audience. Um, and our ability to find, uh, images, uh, that are relevant to that audience, uh, that we can put through our system and generate our own data around. And so we found that in very abstract use cases like, um, the, the example in, in China, which is traditionally inaccessible to most market into most ad targeting, um, there is a way, um, and it's just looking at public information and that was something that was really important to me when, when our core team architected this system is, we said, look, anything that we're analyzing any kind of insight that we are giving our customers.

Jehan:

Um, we aren't doing anything that shouldn't be done, um, in the way in which we, we collected, you know, our, our method is less about kind of studying the individual and more about studying the school of fish. So if you can understand where the school of fish is going, you can predict how members of the school of fish will ultimately react and behave. So it's, it's a, it's been a fascinating problem. Um, and you know, for us just cause we're a pretty lean team. I say, you know, one of the things we learned for sure is focus, focus, focus. So, um, doing, doing a lot, uh, with, with again, CPG, um, uh, primarily, um, but you know, you can insert company name and if they have a visual brand and are doing, uh, you know, commerce over the internet, there's a use case for visit.

Rob:

Yeah. I will say one place that you might want to go in the future because there's certainly a market, at least in my house when somebody gets dressed in the morning, just have your AI say, [inaudible] billions

Peter:

Of images and you are ugly in

Jehan:

You aren't the only one to have suggested that it's best that I, you know, even when, um, when we were raising our, um, our last round of financing, uh, I had a, uh, a really interesting piece of feedback that said, you know, you were going to come in and do this pitch all over again. I would say I've literally designed my entire outfit, my entire image to be specifically calibrated to an early stage software and investor that has an interest in passion in AI. And, you know, like they listed the Midwest and, and I said, well, we'll have to do that next time, but, uh, suffice to say, uh, they're now investors. So it,

Peter:

So, uh, so speaking of roadmap, uh, if you can, uh, as a, as a final question, is there, are there any glimpses of the future, uh, that you want to leave our listeners with before we head out?

Jehan:

Well, I can't get into specifics, um, but I can say that visit is going to make some major announcements in the next couple months. Um, we're, we're on fire and having a lot of fun, and we're finding that our, um, our technology and our platform is, is impacting and sort of finding, finding a home in multiple corners, uh, of the enterprise. And so from our, you know, as we kind of think about the solutions that we build and the partners that we work with, you know, we're, we're driving and we're, we're, we're kind of being driven by our customers. That's really the most two values that really summarize visitors, if you will, heart be human and we're customer obsessed. Uh, so everything that we create is, is really inspired by our, our customers. And so when you think about the ability for, um, a brand to be able to, you know, see through the eyes of their consumers in real time, anywhere in the world, any type of content they're creating, there's near infinite possibilities. So actually a real challenge to us has been prioritizing which things we wanted to do first. Um, but I have to say what we're working on with you guys is, is very, very exciting. So I think that the merging of, of kind of digital shelf and, and, um, understanding kind of how to build and create world-class experiences, um, has as a fantastic compliment with visual understanding and visual brand performance. Oh yeah.

Rob:

One of the most common things we get asked is like, how do you, how do you tell a good image for a bad image? It's easy. It's easy for us to do things like, is it a good title or a bad title? There's just a lot of data and pattern matching and the math isn't that difficult, right? But the images is, is a tough problem. And it's impressive what you guys have done to, to solve it. It's a big one. And as you're saying, there's all this data, we've done our own consumer research, and we've done our own testing where you look at what a computer, what a consumer does on an e-commerce page. First thing they do almost always was the mouse over the image gallery. And we'll just flip through it almost like even if they don't stick around with games, even if they scroll down and they want it, they read about the features functions. Even if they click on a video to watch the video, even if, you know, whatever else they do, they, their eyes go left to the image gallery. They'll, they'll go left with a mouse, the mouse over the whole thing, even if they don't click on it. And it's just a, yeah. So what, what's good matters there, and it's outstanding that you guys have made progress on that difficult problem. So thanks so much for sharing today with everybody.

Jehan:

Oh, you're very welcome. Thank you for having me. It's been so much fun and, and you know, a lot of learning and, you know, we're, we're here because of our team and our partners and, and, uh, you know, folks like you that take the time to, to, to meet with us and learn about what we do and what, what we hope to achieve. So it's, it's a real pleasure. I really appreciate it

Peter:

When, um, when, uh, the, your big news comes out over the next couple of months, if there's a reason to have you back and talk more about it, uh, we'd love to do that. It's really been great to

Jehan:

Have you. Absolutely. Thank you again.

Peter:

Thanks again to Jehan for sharing his case studies with us. Please share this podcast with your colleagues, leave us a review wherever you listen, and thanks for being part of our community.