Harnessing the Power of Imagery on the Digital Shelf

Jun 8, 2021 3:00 PM4:00 PM EST

Request The Full Recording

Key Discussion Takeaways

People are consuming content faster than ever, which presents a lot of need for new content. But there's concern about whether the content is performing well — and for what audience on which platforms? Imagery is one way brands are edging out their competitors. But understanding and measuring the cognitive elements of design that drive consumers to engage, pay attention, and convert has been a struggle — until now.

There's an opportunity to create a competitive advantage by understanding uniquely what works for whom and where. So, how do you carry out these high-level image analyses to harness the power of imagery? What are the steps to boost your product performance across digital shelves?

In this virtual event, Aaron Conant sits down with Jehan Hamedi, Founder and CEO at Vizit, and Adam Colasanto, Director of Consumer Intelligence at Vizit, to explain the power of imagery on digital shelves. They discuss the limits of traditional image analysis, visual brand performance across direct-to-consumer outlets, and case studies of how brands use creative innovation to outperform competitors.

Here’s a glimpse of what you’ll learn:


  • Jehan Hamedi and Adam Colasanto explain how to understand and leverage consumer visual preferences to create a competitive advantage in selling your products
  • Driving more conversion: a new path of creating content with visual images
  • Using creative innovation to outperform your competitors
  • The power of analytics to minimize assumptions in content creation
  • What is visual brand performance?
  • Case study: enabling eCommerce teams to track and monitor the digital performance of your brand, competitors, and categories on an ongoing basis
  • Vizit’s strategies for generating predictive data
  • What the image analysis of various eCommerce categories says about their performance across platforms
  • Should you use the same imagery on Walmart, Target, Amazon, and your direct-to-consumer site?
Request The Full Recording

Event Partners


Vizit helps brands measure, understand, and optimize the impact of their visual content using our Visual Brand Performance Platform.

Connect with Vizit

Guest Speakers

Jehan Hamedi LinkedIn

Founder and CEO, VIZIT

Jehan Hamedi is the Founder and CEO of Vizit, the world's first visual intelligence company. Its visual brand performance platform helps companies measure, understand, and optimize their visual brands to drive more sales, engagement, and connection with their target consumer audiences. Vizit is powered by patented visual AI technology and a proprietary database of over one trillion visual cues that influence consumer behavior.

Adam Colasanto LinkedIn

Director, Consumer Insights, VIZIT

Adam Colasanto is the Director of Consumer Engagement at Vizit, a visual intelligence company designed to measure, understand, and optimize graphic performance. He has over a decade of experience working and optimizing the digital space. Adam previously worked as an Analyst for Paychex, Supervisor for GEICO, Customer Success Manager for Crimson Hexagon and Dynamic Yield, Senior Manager of Market Research for Edelman Intelligence, and the Director of Measurement and Insight for ICUC.

He graduated from Niagara University with a degree in English literature and is an Inaugural Speaker at Google Squared Program (US), where he speaks about leveraging social media data.

Aaron Conant LinkedIn

Co-Founder & Managing Director at BWG Connect

Aaron Conant is Co-Founder and Chief Digital Strategist at BWG Connect, a networking and knowledge sharing group of thousands of brands who collectively grow their digital knowledge base and collaborate on partner selection. Speaking 1x1 with over 1200 brands a year and hosting over 250 in-person and virtual events, he has a real time pulse on the newest trends, strategies and partners shaping growth in the digital space.

Event Moderator

Jehan Hamedi LinkedIn

Founder and CEO, VIZIT

Jehan Hamedi is the Founder and CEO of Vizit, the world's first visual intelligence company. Its visual brand performance platform helps companies measure, understand, and optimize their visual brands to drive more sales, engagement, and connection with their target consumer audiences. Vizit is powered by patented visual AI technology and a proprietary database of over one trillion visual cues that influence consumer behavior.

Adam Colasanto LinkedIn

Director, Consumer Insights, VIZIT

Adam Colasanto is the Director of Consumer Engagement at Vizit, a visual intelligence company designed to measure, understand, and optimize graphic performance. He has over a decade of experience working and optimizing the digital space. Adam previously worked as an Analyst for Paychex, Supervisor for GEICO, Customer Success Manager for Crimson Hexagon and Dynamic Yield, Senior Manager of Market Research for Edelman Intelligence, and the Director of Measurement and Insight for ICUC.

He graduated from Niagara University with a degree in English literature and is an Inaugural Speaker at Google Squared Program (US), where he speaks about leveraging social media data.

Aaron Conant LinkedIn

Co-Founder & Managing Director at BWG Connect

Aaron Conant is Co-Founder and Chief Digital Strategist at BWG Connect, a networking and knowledge sharing group of thousands of brands who collectively grow their digital knowledge base and collaborate on partner selection. Speaking 1x1 with over 1200 brands a year and hosting over 250 in-person and virtual events, he has a real time pulse on the newest trends, strategies and partners shaping growth in the digital space.

Request the Full Recording

Please enter your information to request a copy of the post-event written summary or recording!

Need help with something else?

Aaron Conant

Co-Founder & Managing Director at BWG Connect

BWG Connect provides executive strategy & networking sessions that help brands from any industry with their overall business planning and execution.

Co-Founder & Managing Director Aaron Conant runs the group & connects with dozens of brand executives every week, always for free.

Schedule a free consultation call

Discussion Transcription

Aaron Conant 0:18

Happy Tuesday everybody. My name is Aaron Conant. I'm Co-founder, Managing Director of BWG Connect networking knowledge sharing group with 1000s of brands who do exactly like, do exactly that we network and share knowledge share together to stay on top of the newest trends, strategies, pain points, whatever it might be that shaping the digital space as a whole. I connect with 30 to 40 brands a week, just to do a digital strategy session, but also do a lot of help with partner selection. So if anybody wants to put time on the calendar, I'm always up for a great, you know, chat on digital strategy that's across the board. It could be Amazon, it could be direct to consumer, it could be performance marketing, or international expansion, whatever you whatever it might be more than happy to put some time on the calendar. Also, we did launch a talent division. I know that's been coming up on a lot of calls right now. So if you're looking to hire anybody, we have a talent division can tap into the network of 100,000 plus people and you know, find great people if you're if you're looking to replace any talent that's seems to be around musical chairs right now. The last thing as we get started here is we are we want to get this as informational educational as possible, possible at any point in time. If you have a question, drop it in the chat the q&a or email it to me, Aaron Aaron@BWGConnect.com we'll get it answered. So kicking it off here, three to four minutes after the hour, and just you know, we're going to wrap it up with three to four minutes to go. So we'll give you plenty of time to get on to your next meeting without being late. And, you know, that's pretty much all the stuff to get started. I wanted to jump into this conversation, you know, as they noted in talking to, you know, 30 plus brands a week and content in multiple ways. I think anybody that's been on any other, you know, content driven webinars that we've done is a huge pain point for a lot of people right now. content being consumed faster than ever. There's a lot of need for new content, fresh content, but also there's a side of content, is it performing or not? And how well is performing? What's adapting? And how do we not spin our wheels? How do we get the most out of what we have? How do we how do we let the executive team know what's working, what's not? And then in the end, how do we use it to drive sales. And so we've got great friends, you know, partners of the network over at Vizit. And they've agreed they've, they've come recommended from multiple brands now, and they're doing some really neat stuff. So I thought we would have him on the line today to kind of help educate us around Hey, not just what they're doing in the space, but what does it take to win now in the content space as a whole. And so, with that, you know, I can kind of kick it over, you know, kick it over to Joe first, I guess we can kind of go around the horn here with some quick intros. You know, Jehan, if you want to jump in, that'd be awesome. And then we jump to Adam and anybody can jump in after that. I'm gonna keep letting people in the room. But if you guys want to do a brief intro on yourself, and Vizit that would be awesome.

Jehan Hamedi 3:01

Yeah, sure. Hi, everyone. Thanks so much for joining. Thanks so much to Aaron and bw BWG for having us super excited to hopefully teach you all something new today about kind of how technological advancements and the understanding behind data that's out there and consumer behavior now makes content and visual preference understanding really more possible than ever. I'm here with my buddy Adam Colasanto, who I'll introduce his lead, introduce himself as well.

Adam Colasanto 3:33

Hi, everyone. Sorry, I didn't know you're passing it over to me. But thank you for the time. And we're excited to walk through this. For the sake of backgrounds, just briefly, Jehan and I actually had the pleasure of working together in a past life. So excited to carry on the friendship and the professional friendship as well. But very much my backgrounds rooted and text analytics, insights and innovation, social listening, led research teams and traditional research with qual and quant studies. But very much all of that is rooted in the the voice of the consumer. And what we're doing here at Vizit, which we'll we'll dig into is the view of the consumer. So kind of a new approach to how we're assessing content.

Jehan Hamedi 4:17

Yeah, totally. I think it's really interesting to me as someone also my whole career, I've only spent time trying to understand consumers. And really quickly to introduce you to Vizit we are really allowing for the measurement of content in a new way that is truly understanding of kind of the heuristics behind how humans consume content today. We're all consuming content so differently, and there's so much data out there on imagery that allows Vizit to be predictive of how effective an image and its design can be. For every, most importantly, increasing sales on the digital shelf, which we spend most of our time doing, but really beyond that, making better decisions or eliminating some activity from the conversation and really getting to the right idea faster and more efficiently in a way that creates competitive advantage with your visual brand. And I think this is something that has never been able to be understood in my world. We're in a, in a world of surveys or qualitative inputs, imagery is often left behind because it is so cost effective to produce it is often left in subjectivity. And really excited to teach you all today a little bit more about how we've been able to understand what we have through technological advances. But also, kind of understanding the data around how people buy today and in a unique way, will then kind of jump into some really cool insights that Adam was as prepared to share for us, I think I'll try to be quiet sooner than later. So you can get to that fun stuff, uncover some of those insights, and then give everyone some time for some q&a. But

Aaron Conant 5:56

please don't jump in there just on that note, because we've had a bunch of people join at any point in time, if you have questions along the way, you know, just drop into the chat, drop them into the q&a, whatever, you know, you can email them to me, Aaron, Aaron@BWGConnect.com. But yeah, I'm super excited from the standpoint of what you were just talking about is true is the subjectivity of content, and what's working and what's not, and producing it and the cost of it, and is it working? Or is it not working? And then, you know, everybody's being crunched on budgets, and this year, more than ever, they're gonna be asked to justify where the money went, and was it useful or not? And why or why not? And so anyways, I'm just, I'm excited. But I also wanted to add it dropped questions as we go into the chat or the q&a or email them to me. So

Jehan Hamedi 6:41

yeah, please, let's have some fun with it. So I think it's hard. Probably, it's a hard setting. We've all been in enough of these webinars. I can't wait to be in a position to buy everyone a cocktail instead of hanging up the phone with one of these. But really, just to wrap up the introduction to Vizit and get to the fun stuff about really, what what, how can consumer visual preference be understood and leverage to create competitive advantage of selling your product? What Vizit’s done, we're really on the forefront of AI, everyone tends to be an AI company these days, but we're really on the forefront of deep learning really heuristically understanding the cognitive elements of design that drive consumers to engage pay attention, convert. Um, with that said, the kind of world we're in today sets us up for that, Adam, I think you're in charge, I don't know.

Perfect, thank you, I'll just give you a wink or something. I'm really the world we live in today is why this hasn't been done. There hasn't been ways to understand visual preference, because of kind of how costly it has been to understand these types of things that traditional methods, but it's really, in addition to the advancements in deep learning and AI technology and understanding people the fingerprint of visual preference and cognitive bias that exists. It's also kind of coming back to the availability of data, that we now have trillions and trillions of visual data points and engagements with imagery on the public Internet that allow us to be incredibly predictive of what people react to, you know, when the endless scroll we live in is the biggest reason that visual preference can be so predicted, or it's so easily predicted these days. And not that I guess it's that easy that there are so many people doing it, but it's really what gave us the the idea to go out there and understand what are people choosing to interact with? And Ogilvy's creative global creative director said this a couple months ago, and it's been something that's kind of stuck with me. And I owe him a cocktail for this as well. And it's, it's that the average human is scrolling the Empire or excuse me, I always get the monument wrong. It's the Statue of Liberty every day. It's about 300 plus feet of mobile scrolling, this behavior, this constant scroll of what we choose to interact with, is creating visual bias. It's biasing, biasing us to different design elements that make it extremely more probable that we're going to engage with other imagery that take those types of subtle approaches to design. I don't have to tell anyone on this call, probably how important imagery is it really paves the path to purchase and the human brain is processing imagery 60,000 times faster than text. That's really how that endless scroll the collection of imagery we see every day is is drastically biasing us to what we are having that propensity to interact with including the your hero and carousel imagery on PDP pages. We've really eat with Our eyes, you know, we consume with our eyes, I'm always in trouble because my wife, as me occasionally participate in the family instacart order or channel of choice, you name it. And I'm always ordering things, I'm in trouble for it, because it is a cognitive bias that exists, the snack foods from my childhood always end up in there, I'm visually biased to those brands and those visual brands that have impacted my life. My wife doesn't appreciate that, because often those are cavity inducing brands that I love very much. And it's that bias that that measurement of the cognitive bias of what people are apt to react to at massive scale, that that is allowed Vizit to really help brands with that challenge of creating competitive advantage with each visual that you kind of put out there on on the path to purchase your entire visual brand, if you will.

Aaron Conant 10:55

Really, I just want to jump in here quick, then, you know, So traditionally, when I was, we want to make sure we get onto all the information is, you know, as the brand side, you know, I would I would try an image. And then I would try different image and I would be AB maybe ABC testing, see which one works, I have to wait for click through rate, I have to wait for the for conversion rate to then crunch through that data to see if it really worked. Was it ads that I was driving to it or not? They? What's the next? I mean, maybe we're gonna get to that it's kind of this, what's the next level of that where I don't have to do all of that over and over again, I don't have to wait a month or two months to get that data back. And then try to justify whether it worked. And oh, by the way, in the meantime, there was Prime Day which skewed the results there was, you know, Walmart and Target matching Prime Day or q4 or, you know, whatever else it was it you might say, Hey, we're going to get to that. And but you know, that's top of mind for me is what does that new? What does that new path look like?

Jehan Hamedi 11:53

Yeah, totally. That's a really good point, Adam, you can jump ahead that slide. And and, you know, I come from a different world of measurement that never applied here to which is traditional survey based learning. And whether it's that or an AV test it there, there, there are flaws in either direction, cost and speed of the business, when it comes to surveys, and needing to make a decision before the data is there to one of your points. And then an A B test, which effectively tells you if you're running from a bear Just don't be the slowest piece of content. We do not know that what we are flooding in those examples is good. So the answer is true. Pre testing style understanding of all of our content, including our competitors in near real time so that we understand where we over index, we're going to create opportunity. And really from an eCommerce perspective to reinvent to most of the folks on the calls world that's nearest and dearest their heart. Digital shelf 1.0 is about do we have the right number of images? Do we have enough images? Do we have? availability? Do we have a competitive price that you know, there's a lot to think about but those were generally checkbox best practices, in my opinion, the 2.0 is, are we truly competitive advantage with things we can control, like how we photograph or package our products. And that requires really for the creative innovation side of it requires that ability to measure everything within Khosla without cost limitations to know how we're performing relative to competitive content. And, and be statistically advantage before we have to fly a bad piece of content alongside with what we like. or frankly, waste time spinning cycles, waiting for a CMO to say I like that one that's equally you know that I don't know what's worse, not having data bind a decision, or spending time and energy to not have data by a decision. But we there's a fair amount of both going on. So Adam, you can jump ahead and we'll get to the fun stuff. I'm still on too much of the time before we get to insights already. But what what this comes down to what this ability to understand someone's visual world and the bias they have and measure imagery in such a way that we can have that statistical confidence that we have a better image, and we will sell more product with a better image is what is what our research has shown. If you have a better image at each of those inflection points on the path to purchase those, those kind of visual brand journey your customer goes on, you will win whether it's the earliest awareness stuff, or the hero image that is better than your search that your core competitors as you view them. If we have a better piece here, we will sell more product and really understanding. You know, search is so important that to your point earlier like this is something that everyone's thinking about every day. But as that dynamically changes every day, as as the content we're flooding and our competitors are fighting every day changes. It's something we really need to have be on the pulse of and that's not something that you know, an AV test. Or, you know, millions. And the budget required to do that in an ongoing way with research is ever going to be possible. So that's really what we set out to do as an organization is allow for this real time measurement and highly predictive ability to understand what's going to move your customers to buy, all the while kind of answering these core questions in terms of like, what are those design elements, it's things like shadows and angulations. And and contrasts that are having that impact on the on the norlin, logical side of the visual buying process. It isn't things like put a dog in your image, although everyone likes a dog, and it certainly could be. But it's, it tends to be those subtle things that don't disrupt creativity, they don't disrupt the creative process or stifle innovation like I did when I was measuring an ad in my past lives. It's this really cool empowerment, the creative process allows you to get to the idea faster. And and then once you have all this content, and I know we all have a lot of content, it turns out, most brands have more than they know what to do with they don't know what works for whom where and and that's a big part of kind of where we're we're solving some problems, too, is just that ability to know, what is my best foot forward. And there are often times where you don't need to create new content to be the most effective visual in a given application, whether that's a carousel image, or, or something on our way to represent your product and a hero image. But at the end of the day, what we've found in the way that everyone needs to be thinking about this is having a better image than our competitors in those key moments will drive success in those. But most importantly, sell more

the content piece, in terms of kind of how your consumers see you goes far beyond just those, those elements I mentioned today, you know, it's every visual representation of your brand. And it you know, the, the process of creating a piece of content can be so much faster. Well we understand the trends. And the elements of a piece of content that do resonate with the consumer, maybe it is those shadows, maybe it is those angles, and being able to compare in real time to the entire competitive set on a channel, or more broadly, all the visual touch points of your core competitors. And in a given category, we can really understand and monitor over time, those changes we discussed, you know, where am I stacking up in relative to search rank on Channel x. And when something changes, and we lose a couple points on a channel, oftentimes, it's a content piece that's triggering a much bigger reaction organizations, redesign brands changing packaging, when when we just haven't had a blind spot to content and the measurement of content that is forcing us to make kind of reactive decisions rather than monitor to put our best foot forward, recognize where we can fill the gaps with things we've already done. But most importantly, use everything we've already done to ensure we never make the same mistake twice. And that we're kind of innovating on how we present ourselves.

Aaron Conant 18:16

I mean, I think another takeaway that I have here is guiding the content team on what to create, right, because we're leaving it kind of open to almost guessing we don't have to write, we don't have to leave it open anymore. We can provide them, Hey, this is what's resonating things that look like this. And then over time, they're going to learn and get better and better and fine tune it right now. Every single piece, it's you know, without data analytics, it's kind of like a guessing game.

Adam Colasanto 18:47

That's exactly what we're our mission, our value prop is really to, as Joe mentioned before, remove subjectivity out of the equation. So you are making more data informed decisions around what is actually resonating with your, with your target consumer. And we actually have a nice slide coming up and like I think two or three that actually showcases how that evolution can play a factor with assessing your own content category, and then using those to fuel your your own creative teams to build the right assets. So that's absolutely something we're addressing here. Joe, anything else you want to say before I take over and jump in? Let's get

Jehan Hamedi 19:23

to the fun stuff. I always think the most exciting part is sharing some insights. And that is Adams realm. So I'll get out of the way. Perfect.

Adam Colasanto 19:31

So we're super excited to talk about what we're doing over at Vizit we've invented and created this visual brand performance platform, which is ultimately an algorithmic understanding of your true quote unquote, view of the consumer. What we're doing is helping brands measure, understand optimized visual content designs, for their specific consumer types. But most importantly, it's all happening in real time. So thinking about those traditional search rays are a B tests where you have to invest not only the money, but also the time to get those those returns back. We're doing this in a matter of seconds, where you're going to be able to understand what's effective and what isn't. So I'll walk through a couple examples here about eCommerce packaging, Hero imagery, carousel imagery across a couple different CPG categories, and how they can be measured and optimized for different consumer types with some real data and some real examples. So visual brand performance, it brings an AI driven real time visual test and learn to every single stage in the digital shelf optimization process. And this is what we were just talking about. It starts with understanding your own content and the images you already have in your dams, or Pam's in your content experience platforms, really giving you the sense of of all the content that I already have in my fingertips? What's actually good, what's actually going to move the needle, and where can I kind of build upon from there. Once you've analyzed your own content to assess what's effective, and what isn't, then we can look more broadly to analyze analyze an entire category, or specific competitive sets to identify trends and themes. And that will inform how you're standing out within the category for your ecosystem of products. And that's where we can take those learnings, those trends and insights. And you can inform your marketing and creative teams with that visual data and feedback that they need so that they can then design highly effective content. Additionally, what we're doing is we're actually simply supplementing and or validating traditional research methods with a lot of our datasets, we've been very predictive of not just us based, like focus groups or surveys, but also focus groups in Beijing, or we did a really cool analysis on motorcycles in France and Germany. And we've been predicted completely predictive of those results and insights as well. The final step in this process is enabling your eCommerce teams to track and monitor the digital performance of their brand, their competitors and their their category on an ongoing basis. And this is hyper important because there's always going to be new content created, there's going to be new products, there's going to be refreshes there's going to be rebrands. So understanding that evolution and natural ebb and flow of consumer preferences based off of the new content that they're engaging with, it's it's hyper important to stay on the cusp of that. So you know, you're always on the cutting edge and creating content that's relevant to your your consumer types. So with that said, we actually have an example today within our brand, our brand platform, joke, you give me a thumbs up, if you can still see this. Okay, awesome. So this is deep within our Vizit platform, we are in a sub folder. And what we did is we pulled down the top 100 brands for cleaning spray on Amazon, and we pulled 55 product hero shots. So within this folder, we're looking at all the different brands for cleaning, spray your images on Amazon. When I assess our different audience lenses, which can be seen here, we can do retail specific ones, we can do interest groups, specific ones basic demographic, when I click millennial men here, all of the images of those 55 will be sorted and re filtered. So that we can start to identify even at the very top level some some surface level trends to see what's popping and resonating with with specific cohorts. So right off the top, we see that it's really bright colors method appears to be a top brand with their kind of bright fluorescent colors within the top six here. Contrast that with women 18 to 24, we're going to see a completely different dynamic and preference, which will be more multi bottle much more traditional in the sense we see Lysol and Clorox and Mr. Clean appearing. So just at the very scratching the surface, we can already see that there's a huge skew in preference even looking at different just male versus female and age demographics. When we click into a specific image, so I'll do this lifestyle one, I have a number it in a different tab. This is what we call an image scorecard work, we are going to get the Vizit’s proprietary score to understand how effective this image is. And then benchmark data. So looking at the entire category, again, that top 100 Amazon cleaning products, we can see that this image specifically is 22.4% better than the average hero shot for cleaning spray on Amazon.

utilizing our heat mapping technology, we can actually drill in and identify what's actually driving that performance and what's resonating with women 18 to 24. And super interesting to see that it's actually the citrus that that lemon fruit that's on the bottle itself that's really driving a lot of effectiveness and performance with this specific image. So just a quick real life example that we wanted to showcase. But jumping back to the presentation, we did analyze an entire product category for both snacks and coffee machines on a number of different retailers respective of different audiences that were retail driven. So we'll start with snack after analyzing hundreds of the top search net products across Walmart, Target and Amazon. we summarize the findings here. And some there were some very clear visual trends. At the retailer level, we can see that Amazon or excuse me, Amazon, Walmart has the highest overall visual effectiveness rating driven by both hero shots and supporting image content. interesting anecdote, Walmart also had the highest number of images per listing at 6.17 as compared to targets three point excuse me, 5.34, and Amazon's 4.39. When we look at retailer specific so on Walmart's digital shelf, we can see that when it comes to product hero imagery, Pringles and Nabisco products have the highest scoring individual product hero images, as well as the highest scoring carousel imagery for Walmart shoppers. Well my team does is surface a lot of these trends and insights and recommend those strategic tactics for you to implement in your own content creation. So some of the insights that we surface was for hero shots. Pack rotation, so packing at a slight angle, maybe about a quarter rotation is noticeably more effective for Walmart shoppers, out of box out of box presentation. So visuals of snacks within cardboard boxes, including like an individually packaged snack in the foreground are incredibly effective. And Pringles just the packaging as a whole was noticeably more visually effective for Walmart shoppers than any other snack product packaging design. We look at the supporting carousel, snacks within recipes and this really detailed texture. So getting a close up shot or being a part of a recipe was incredibly effective for Walmart shoppers, snacks on the go, where you see somebody kind of tucking that snack until a book bag or a purse or bag of some kind. And then snacks on shelves. actually seeing the snack displayed within a cabinet or some kind of shelving unit was incredibly effective for Walmart shoppers as well. Looking at Amazon digital shelf, the products packaging, the hero shots that featured ample whitespace was incredibly more effective for Amazon shoppers, images that had multiple bags or products or kind of a fanned display really resonated as well. And then another trend we saw was isolated produce imagery, it was incredibly effective. So we can see the pineapple, the oranges and the grapes and the peppers, they're thinking about the carousel imagery, overhead angles. So there is a very clear advantage for Amazon shoppers when snack food was was photographs photographed from an overhead perspective. Bright colors with that generated, you know, contrast with the product also generated higher scores with this particular audience. And then forest and kind of outdoor scenery. So specifically with Jeff links, they're supporting carousel really represents this kind of outdoorsy camping type field. And that works really, really well with Amazon shoppers.

Aaron Conant 28:13

Yeah, want to jump in here as a question comes in and just reminders to others. If you have questions, drop them in the chat there or email them directly to me Aaron Aaron@BWGconnect.com we'll keep getting an answer. So a couple different ones have come in. One is how are you quantifying this? It's a thing. Great and panel eyetracking tea leaves.

Adam Colasanto 28:36

very fair question. And we can go pretty deep on al keep it kind of high level at the moment. But this is a true artificial intelligence play, what we're doing is assessing hundreds and 1000s of assets relative to your specific consumer lens. And then within that image, there are 10s of 1000s of triggers or attributes that on a big grand scale, we aggregate into trillions of data points, what the algorithms are doing are now identifying the trends and patterns within that imagery that resonate with your specific consumer. So that when we go to test this image of honey graham crackers, we can now say, Alright, based off of what we know resonates with an Amazon shopper lens or consumer, these specific things work well. We can draw correlation. And now we've taken that aggregation of data points, and just the output is a very simple, easily digestible, zero to 100 score that is likened to appeal. Jehan, how did I do? You’re on mute, buddy.

Jehan Hamedi 29:42

Always was really great Adam II it's like it's like you work here. I know I the simplest way to put it is it's massive pattern recognition driven by behavior. This is behaviorally driven prediction rather than the stated behavior as compared to like survey. This is the bias you see when you are in you're sitting on your couch at the end of the day going through that scroll, what is it you choose to follow? What is it you interact with? And what we find? Is there a massive pattern to those 10,000 elements and have unique features that aren't the things someone would tell you they like in the y of a survey? It is those shadows and angles and subconscious elements of design, that it allows us to uncover by having such massive data sciences and pattern recognition capability.

Aaron Conant 30:29

Yeah, and this is the behavior click through conversions, sales glance views.

Jehan Hamedi 30:34

Yeah, that's something we've, we've really found were predictive of both the answers in the surveys, as Adam said, but also of any kind of action, you know, I really think of our score our simple single number as kind of a composite appeal, where we tend to be predictive of those types of visual appeal questions. But also the action. You know, every time we've optimized an image for a digital shelf setting, we've seen measurable lift. So it's, it's, it could be click, it's really just kind of an open ended intent to pay attention, which tends to correlate with action.

Aaron Conant 31:08

No, I love it from the standpoint of you know, I wasn't thinking about it that way. But if I'm not looking to purchase something, but an image catches my eye, I might click into it. Now, I'm still not going to pitch, I'm still not going to purchase it. But it means the image did what it was supposed to do, which get me to pause and look at it. Right, which is our Bluetooth key through the constant scrolling is happening. Awesome. And if there's you know, everybody keep dropping in questions. It's awesome into the chat here. We'll keep getting them answered. But I mean, that's, it's interesting. So kind of a collective on the back end. Oh, how are people like analyzing the data today? And what's what's changing.

Adam Colasanto 31:52

It's been a lot of different use cases, our application is incredibly broad. So we've gone to market with eCommerce because we've seen that better imagery sells and we can draw that correlation exactly to your point, Aaron, conversion, click through add to cart, Show me the money with with revenue and sales. And it's an easy way to validate what we what we've been doing. But the application is so broad, it can be putting better content on your social channels, creating better pack design to sit on the physical shelf. We're kind of dispassionate about channel and use case. Some of the more recent client case studies that we've done is to help understand wine labels, and how they stand out against core competitors. We've done some compliance things for badging on hero shots with with some some big soda cab companies, and how they present themselves on both DTC sites and specific retailers. There was a CBD company and they looked at what's the right way to photograph? athletes? And should they be you know, it's a sports cream CBD infused pointment. So should an athlete be playing football? Or should it be soccer or basketball, what's going to be the best way to present our products from a lifestyle perspective that will drive you know, sales and resonate with our consumers? So the application is incredibly broad. And we're kind of just scratching the surface surface with a myriad of different clients and use cases?

Aaron Conant 33:19

Is there there are people using this to then also, I can see it from a package design standpoint. Right? I mean, you can fit it all the way back into Hey, if I'm going to launch a new product, do is you know, using the data and the imagery to actually get a package and or, you know, combo pack, or whatever it is to actually design the actual product itself.

Adam Colasanto 33:45

One of my favorite case studies that we've done is with us potato chips, they had a very outdated, antiquated pack. And more specifically for a niche product for avocado oil, potato chips. So what we did was we assessed I think, was 23 to 30 some odd images, and identify what's the appropriate way to display avocado oil chips, and we did a full pack redesign. And not just like the presentation, but also the brand logo itself. While we're not interpreting text or copy, we are analyzing the composition of it and the aesthetic. So how big should the font be? What What style of font, what color, so on and so forth, the positioning and where, where it is relative on the pack itself. And what we saw was a 55% sales lift, and I think it was about 32% increase in findability on the physical shelf itself, just by kind of informing what the right pack is for a potato chip consumer.

Aaron Conant 34:47

Awesome. Love it. So another question that comes in a couple more here. How many times per year would would you recommend making significant changes to PDP content once data is being utilized to create assets monthly quarterly semi annually annually. How often should people? I mean, yeah, awesome. I mean, I think that's the next level, like, awesome, you know, information here, like, but then how often should people be updating it?

Jehan Hamedi 35:15

Yeah, from my perspective, this is a more dynamic space than ever before, we're finding content changes and your competitive sets more frequently than ever. And, and from my perspective, first and foremost, it just starts with assessing the ground truth. But I in my world, I think quarterly is the right kind of lens, unless you're aware of change, you know, there's, we can be pretty real time about knowing what's out there at any kind of visual inflection point, but specifically speaking to the digital shelf. And I think just keeping a pulse on that, at least quarterly is the right way to ensure you're not at that reactive position of why did we lose a couple points, you can be ahead of that conversation and bring insight to the business that allows you to be proactive as to making changes and having some reason behind it.

Adam Colasanto 36:07

We my POV is is kind of the opposite side of the fence, it's not so much when should we update. But you know, when we talked about monitoring here, you understand what you have you understand what the category in the competitive competitive set is producing. You've done your validation. Now it's just about assessing. So when you start to see those dips in a couple points, you're not dropping a couple million on innovation to try to understand it, you can take a proactive approach. And it doesn't have to be a set rigorous schedule quarterly, we need to be updating our content. It's more about, hey, we're seeing an evolution in preference based off of our consumers, let's make the change to ensure that we're now continuing to resonate and keep that revenue generation up.

Jehan Hamedi 36:53

And I only meant to be aware of what's going on at least quarterly not content, I think the content update is really just about how are we doing relative to competitors? What is changing around us? You know, no, I

Aaron Conant 37:06

agree. I mean, if it's not, I mean, I'm kind of processing this, as well as if it's not having it's still see if you will see you go through the optivisor content, maybe go six months, maybe it goes nine months, and you're still at the top, like then don't touch it. Now you have a justification around, hey, don't deploy assets to update something that's working. Just because, you know, it was a we're supposed to do this twice a year. And now it's time to do it. You know, it's not like just Google sitting there, you know, looking for content refresh. This is is the content engaging? Is it working? If it is don't touch it? It you might actually save yourself from hurting sales? No, so no. Love it. How it is? So another question that comes in. It sounds like kinda like Adam and Jehan, that's what you were saying? You're kind of getting that that as well. Right? Like, maybe don't touch it.

Adam Colasanto 38:00

Yeah, exactly. But it's the it's the understanding, Never before has it been prudent, like cost effective, to be able to understand how content is performing relative to the, you know, the industry as a whole, and who your consumers are. And now you can do it in real time. So what's it?

Aaron Conant 38:20

So the benefit? So the benefit here is real time. So right? How do we know using this service versus, you know, the A B testing that I mentioned earlier in the call? How do we know that this is better than that? Is it the real time? We're

Adam Colasanto 38:34

not we're not trying to replace anything I see us and Joe, you can obviously disagree here, but very much complimentary to, you know, what a lot of companies do to validate success, we're not, we want to become a Stage Gate in the process of validation of what the right content is to push and syndicate. When we talk about a B testing, you know, we're not thinking about Joe, it's a perfect example, when you're running from a bear Don't be the slowest one, right? Like it's a single variant test. We're doing true multivariate testing with artificial intelligence to give you a full assessment of what that asset looks like, and what's working and what isn't. And the speed, and the cost savings of being able to do that is what's really separating us. And really, it's the first of its kind, but that's what's giving us the notoriety right now.

Jehan Hamedi 39:22

And I think like, that's, I'm totally with Adam, but we're also where it changes the entire workflow of the business is really important. How you develop content, where your creatives are starting is in a better place where you can assess everything and understand the trends that are working, and what people are resonating with. Cuz then you're really bringing even if you use an AV test at the end of your cycle to validate you still at some point had to get to the confidence in those two assets to take to the AV test. We're now going to eliminate any kind of, you know, kind of loss of impact of a bad piece of content. Because even because we are now bringing best, the best to to a single variant type of analysis, all the while having allowed creatives to test and learn and get to the right idea so much faster than would have been possible. And that subjectivity time we talked about that, but it again, I think it annoys me more that it takes time to be subjective than just the fact that we are subjective in any case. These are great questions though.

Aaron Conant 40:27

Yeah, no, I keep coming in overall for over all for better for you brands. Are you seeing that consumers want to see very visible diet made I'm thinking this is getting to, you know, get very specific, around just insights as a whole. In the better for you brands? You see in the customers want to see very visible diet free, free from diet slash free from call outs on packaging, for example, to customers, both conventional and national want to see a bold gluten free cut, call out paleo, no sugar added example like that, or should those call outs be minimized or not shown because they can be detrimental to the conventional customer? I don't know if you have any insights there?

Jehan Hamedi 41:06

Yeah, that's a fun one. It's fun. We do a ton of research on theirs. It would be annoying research answer, it truly does depend on the unique lens in your consumer, we're essentially teaching computers to see the world as you view your segmentation. So it allows you to really make your understanding consumer a weapon for the business. Where what as an example, target shoppers target is pretty famous for not allowing a ton of text on imagery in the econ environment. What we find is even Gen population groups like an Amazon buyer and a target buyer have different enough makeup and have different cognitive biases as a whole. That things like using more text on one channel versus the other is certainly an insight we've surface on a repetitive in a repetitive way it comes up all the time, things like how do I position the USDA logo and badging? It's about where does it sit on the pack more than anything, because of the colors and contrast of the artwork that creates the backdrop. And then I think what we found is text matters, it will always matter. It matters more to some buyers, but our biggest thing is telling you where to most effectively put it to garner the attention that it deserves. And then, you know, aligns with the rest of the research and your innovation as to what are those most motivating claims? And let's put the right one and the most powerful spa on pack. It was perfect. Yep.

Aaron Conant 42:29

Awesome. Quick question. And you can put this if you if you have to, or the brands, including this presentation clients Vizit, so core hobbies, etc.

Adam Colasanto 42:39

We just did a category assessments, we just went to each individual retailer and just top 100. So it's the analysis here is not even about if they're a client or not. It's just understanding this specific category. So again, we're trying to understand categories as a whole and how different brands and products kind of play off of each other. What's working, what isn't. So it's it was this compassionate whether they're a client or not?

Jehan Hamedi 43:03

Yeah, well, it's our clients are the biggest CPG players in the world. We spun out of Pepsi's incubator a couple years ago, which gave us cool exposure. And it makes us really curious. I think that's our DNA. It's like, let's run this research we have that is the power of this type of understanding is you can test anything, your competitors stuff, what's going on in the world, entire categories, really do broad assessment that you're for us. This is more of a passion for insight discovery, and thought leadership stuff that poor Adam spends all day looking at images as a result for this, this passion. But most of our customers do tend to be the big CPG players though, too. Awesome. Love it.

Aaron Conant 43:47

Are there other things that come up routinely? I don't know if there's other, you know, insights. Okay, yeah, here, we'd love it. I want to make sure we can get through all the info here. But thanks, everybody for sending over the questions. We'll keep getting them answered as we go here. Just drop into the chat or email them to me.

Adam Colasanto 44:01

Yeah, I love the questions. And I love having a dialogue. I have a couple more slides that are insights driven, but I'll keep it brief because I think it's more valuable to talk about what we're doing and how you can harness it internally within your organization. So this is the last slide for the snack category and then we'll talk about coffeemakers just briefly but again getting back to these trends when we look at uniquely target shoppers and the products here images for snacks, packages and products that contain the color blue worked really really well Frito Lay variety pack product here I'm just with kind of a slant are staggering of product are highly effective. And then again, we see Pringles emerging as a top performer for hero shots with the carousel imagery, texture and close up again. So seeing the actual snack food drove a lot of effectiveness for target shoppers hands including Grabbing for the snack product was a great trigger for performance and effectiveness. And then that overhead kind of flatlines heights that tie, excuse me style of image was very popular for this audience as well. So that kind of wraps up the snack category. We have a similar style approach for for the coffee machines. And I'll kind of move through this quickly so we can get back to some questions and excuse me, my cat just jumped on me. I swear every call she always says negative parents, it's like she knows I'm on a video chat. So when we analyze the top search coffee machine product imagery, if a retailer level, again, Walmart has the highest overall visual effectiveness rating, driven again by both hero and carousel content. However, unlike the snack category, Amazon actually had the highest number of images at seven per listing, followed by target at 6.13 and Walmart at 3.5. So just kind of an interesting kind of insight on how that kind of differentiates when we look at uniquely Walmart, digital shelf, and the Walmart shoppers, we can see when it comes to products hero, then vino tab GE and Keurig have the highest scoring product hero shots. While GE their carousel imagery tends to be the highest performing or highest scoring overall, with the product hero shots, images that featured steaming cups alongside the coffee machine score really, really well. Coffee pods that were contained in drawers, rather than on an actual display. Those scored really high with Walmart shoppers. And then compared to more manual machines, images of more modern looking coffee machine tended to perform much higher for the carousel shots. High quality lifestyle images, including countertops, specifically with GE branded ones scored really, really high. close ups of the digital features scored well. And then on like pajak links kind of outdoor camping scenery, those outdoors, the shots performed, actually pretty poorly. They were they were not strong trends that resonated with Walmart shoppers, the Amazon digital shelf with Amazon shoppers, they preferred a coffee pot handle rotated to the left. And this kind of goes back to what Jehan was talking about before. It's subtle, subtle differences of how an image should be kind of shot on a photoshoot. And we're actually working with brands that are putting us setting up a laptop in a photo shoot environment and taking pictures testing in real time, and then making the appropriate changes on their set. We were working with an electric bike company. And literally they had smoke machine going they were testing should it be three bikes in the image or one? Should there be a rider on the bike or no should there be somebody standing next to it. So all these different subtle changes you can do to really understand what's going to work with your core consumer. Steaming coffee cups did not work well for for Walmart shoppers, and products here are images that had multiple beverages surrounding the coffee machine also performed pretty low as compared to Walmart. For the carousel imagery, depth of field and blurred backgrounds were incredibly effective. Again, we see hands doing really well kind of touching the machine. And then coffee cups without the machine present just a standalone, here's the here's you know your coffee and your mug performed really, really well for Amazon. And my last slide for you guys then we can open up for for broader q&a. Target digital shelf with target consumers, stovetop, espresso makers were the highest performing products hero image on target

images of the cup of coffee machines slightly rotated, performed higher than than those that were there straight on. And then the machine itself. If it was just a standalone tended to perform lower, as compared with with multiple that were that were with beverages next to it. And then the carousel imagery cropped focusing on the upper components did really really well. There were a handful of images of women drinking coffee in a kitchen type setting that performed incredible for for carousel, supporting imagery. And then finally, images of people actually swapping out or changing their single use pods were again incredibly effective for target shoppers.

Jehan Hamedi 49:26

Yeah, the insight that we see very frequently that is a fun one point out and I guess a sad one for human nature is narcissism matters and imagery, right like picturing yourself with your with the products and and being able to have that human piece of the puzzle provided where possible. 10,000 big impact.

Aaron Conant 49:46

You know, I want to I think, actually one of the things that I was just, you know, processing was this question that gets it's been fielded for the last I would say two years before because before that won't While trying to play catch up hadn't made enough grounds, but do I use the same imagery? So we set aside bullet points and product descriptions and a plus Detail page content? If we just look at the main imagery. The question out there is Ben, do I use the same imagery on Walmart on target on Amazon? My own direct to consumer site? Do I use all of it the same to keep the brand image the same? Or do I use different is import perform differently? And the answer is, it depends.

Jehan Hamedi 50:40

But it's quite possible in equivocally. Yes, it differently that's every time. And it's not

Adam Colasanto 50:46

just retailers, it's market as well. And we're working with some cosmetic companies that literally That was the question, because they were rolling out content across France, Germany, the US, some South American companies, or excuse me, regions, all the same content. But those aren't the same consumer preferences in each of those regions. So it's both channel specific, and marketplace specific as well.

Jehan Hamedi 51:09

I even like a nerd, just like a target, same mom, target shopper, you know, would have in the Boston and in Lake Tahoe, California would have different pre cognitive bias. And that's what drives it. Right. So there may be some overarching trends, like Amazon shoppers tend to like to see more experiential photos. But for you how you view your brand, you're always going to want to understand the best piece of content for that unique setting.

Aaron Conant 51:39

Wow. Awesome. And then there's a tool to help measure that and dictate What's it supposed to be? No, I love it. And I see that we're right at time here. So we're gonna have to kind of wrap it up. And I'll kick it over to you know, Joe and Adam for some key takeaways. But real quick, thanks to everybody who was able to send in questions. It was great. You know, thanks, obviously, Joe, and Adam, for your time today and being so open for us to throw questions at you. You know, again, encourage anybody via the opportunity. Yeah, have a follow up conversation with them. Take a look at the platform just came recommended from a few different brands and was really intriguing. Because I know a lot of people are dealing with content and and how do we use it? And how do we optimize it and everything else that goes along with it? So worth a follow up conversation with him for sure. And also look for an email from me, I'd love to have a conversation with you see how things are going in the digital side. If there's any other topics or you know, there's any pain points, you're looking to address more than happy to have a conversation with you on it. You know, I think to you, Jehan, first key takeaways, and we'll go to Adam and you know, try to keep it you know, under 30 seconds or so if we can, and we'll wrap up and get everybody out of here on time.

Jehan Hamedi 52:47

Yeah, totally. Thank you so much, Aaron, and the whole BWG family and everyone for joining today. It's so much fun for us to talk about something that we're so passionate about. And the bottom line is content sells, and we've never measured it. And we now have an opportunity to create competitive advantage by understanding uniquely what what works for whom, where, which is always the answer we have found is kind of left behind because of how imaginable image imagery has been in the past. And I think the other key thing is, image is being processed faster than the why I'm by and we need new methods to compliment traditional thinking like this that give us that biased, visceral reaction to an image to create that advantage.

Adam Colasanto 53:32

Yeah, the only thing I would really add because it was very well sajo is we're trying to remove subjectivity from the equation. So you can make better, faster, cheaper, smarter decisions about what the right content to put on your specific channel. So as Jehan said, the right content for who and where.

Aaron Conant 53:51

Awesome. Well, thanks again, Jehan. Thanks, Adam. for your time today. Thanks, everybody who dialed in, just a heads up we're going to kick off in person events in q3. Probably a little bit mid q3 or so. So if anybody's interested if you're in a tier one city and you're interested in joining, don't forget to go check out you know, our Connect events landing page ready putting a bunch out there or just shoot us a note and say, Hey, I'd love to attend if you're in a city near me. You know, it'd be great to get back to rob now we'll have a cocktail hour and in a cool dinner together. So be on the lookout for that. And with that, we're going to wrap it up. hope everybody has a fantastic Tuesday. Everybody stay safe. I look forward to have you in a future event. Thanks again. Thanks, Jehan.

Read More
Read Less

What is BWG Connect?

BWG Connect provides executive strategy & networking sessions that help brands from any industry with their overall business planning and execution. BWG has built an exclusive network of 125,000+ senior professionals and hosts over 2,000 virtual and in-person networking events on an annual basis.
envelopephone-handsetcrossmenu linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram