Mike Graen sits down with a variety of guests to talk about the importance of Computer Vision (CV) product recognition software to take digital images and transform them into on shelf availability issues.
Joining Mike are Anand Prabhu Subramanian (Infilect), Angam Parashar (ParallelDots), Renish Pynadath (Snap2Insight), and Matt Greene (TRAX Retail).
Hello, my name is Mike Graen. Welcome to today's
Mike Graen:podcast on on shelf availability hosted by Conversations On
Mike Graen:Retail and the University of Arkansas Supply Chain Management
Mike Graen:Research Center. We are going to spend some time today talking to
Mike Graen:leading edge product recognition software providers. Whether you
Mike Graen:have a robot that is scanning products in a shelf on a store,
Mike Graen:or you have fixed cameras, or you have a service that allows
Mike Graen:people to go in with their own devices and take pictures. All
Mike Graen:of that is requires a ability to be able to say what exactly was
Mike Graen:seen. Product recognition in store condition software does
Mike Graen:that, and we are we're very fortunate to be able to see from
Mike Graen:the leading product recognition software providers. So let's get
Mike Graen:started with the podcast.
Mike Graen:Well, hello, everybody my name is Mike Graen and it is a
Mike Graen:pleasure to be with you. We're actually recording this on a
Mike Graen:Friday afternoon. So for those of you who are joining us live
Mike Graen:happy Friday. We've got a really, really super exciting
Mike Graen:panel that we're going to spend some time talking with today.
Mike Graen:The topic is continuing to be on shelf availability, and
Mike Graen:specifically being able to use computer vision software product
Mike Graen:recognition about exploring in store conditions. So I'm going
Mike Graen:to go over a couple of kinds of things that we want to cover
Mike Graen:first, and then we're going to introduce our tremendous panel
Mike Graen:who's with us today. The first thing I want to kind of walk you
Mike Graen:through is just a little bit of background on me, I've been in
Mike Graen:the retail and supplier area for over 40 years. I've been with
Mike Graen:both Procter and Gamble, Walmart and a third party merchant
Mike Graen:called Crossmark. And now really work with suppliers and giving
Mike Graen:back to the University of Arkansas and Conversations On
Mike Graen:Retail for information about on shelf availability kind of
Mike Graen:capability. So one of the things that I think it's important to
Mike Graen:know is just kind of what the ground rules are. Conversations
Mike Graen:On Retail is helping us to co host this along with the
Mike Graen:University of Arkansas, we want this to be very interactive. So
Mike Graen:we're going to ask you to be prepared to answer questions or
Mike Graen:ask questions, there's a couple different ways you can do it.
Mike Graen:One way you can do it is after each section of this
Mike Graen:questionings that I've already got for these folks, we'll go
Mike Graen:ahead and come off mute and if anybody has wants to act, ask a
Mike Graen:question live and interactive, you're more than welcome to do
Mike Graen:that. If you'd rather stay and just be anonymous, you can text
Mike Graen:the actual questions to our chat function and when we get a
Mike Graen:break, I will certainly ask those as well. When you're
Mike Graen:asking the question, we'd asked you to keep the video on, use
Mike Graen:the chat function. And then we've got, you know, the obvious
Mike Graen:thing is we're gonna run this under antitrust guidelines,
Mike Graen:we've got, you know, four incredible companies and leaders
Mike Graen:who are all basically direct competitors of them's of each
Mike Graen:other, but they're all in the same basic space. So the
Mike Graen:purposes here is we're not talking about pricing and
Mike Graen:margins, or anything that would look like competitive or
Mike Graen:antitrust violations. We're looking to educate the industry
Mike Graen:about this role that this starts. So a couple of things,
Mike Graen:just just a couple of introductory things that to kind
Mike Graen:of set this up. First off, I want to thank Conversations On
Mike Graen:Retail and I know Matt Fifer is actually hosting this as we
Mike Graen:speak. Matt's a great resource and Conversations On Retail
Mike Graen:continues to provide education for suppliers and education for
Mike Graen:the retail industry around that. So I've encouraged you to, to,
Mike Graen:to follow the Conversations On Retail link on retail to see
Mike Graen:more about that about getting involved with that. And then
Mike Graen:obviously, the University of Arkansas Walton College, which
Mike Graen:is the number one Gartner supply chain university, a part of the
Mike Graen:journey university. We're excited to have them post both
Mike Graen:the audio and video podcast to that. So we appreciate them very
Mike Graen:much. So a couple of background kind of thing. So this all is
Mike Graen:about on shelf availability. And we typically will see the Sam
Mike Graen:Walton quote about making sure you're meeting customers needs.
Mike Graen:I actually found this the other day, this is Doug McMillon,
Mike Graen:who's the CEO of Walmart. And I thought this was pretty, very
Mike Graen:aligned to what Sam Walton was saying, but it's really pretty
Mike Graen:simple. If you're not meeting the wants and needs of the
Mike Graen:customers, you're done. There's not a lot of loyalty there. So
Mike Graen:if customers go in and they want to buy something from a store,
Mike Graen:and the store doesn't have it on the shelf, guess what they're
Mike Graen:going to get the product and they're going to find whatever
Mike Graen:way they can to do that. They're not loyalty to a particular
Mike Graen:retailer, their loyalty to the product they want. And these are
Mike Graen:a couple stats we've shared before, I'm not going to read
Mike Graen:them out loud to you, but basically, if you don't have the
Mike Graen:product on the shelf, or if you have out of stocks or whatever,
Mike Graen:customers are very resilient and they keep you know Amazon. Well
Mike Graen:thank you very much for the additional business of pushing
Mike Graen:folks that way. On shelf availability, and this is why
Mike Graen:these group is so important is you look at a forefoot section
Mike Graen:of, of shampoos and conditioners, this happens to be
Mike Graen:head and shoulders at a Walmart. You obviously see the one
Mike Graen:particular case where you have an out of stock, what you don't
Mike Graen:see is there's a lot of things like missing labels, suppliers
Mike Graen:typically call that distribution void, you've got pricing
Mike Graen:situations where the price at the register and the price
Mike Graen:actually at the shelf are different. And then you have
Mike Graen:opportunities like plugs, and these guys will sure talk about
Mike Graen:that, but that's where you have a label, you have the product,
Mike Graen:but they're different. The product, a product that's there
Mike Graen:and the label is there is different. So as you step back
Mike Graen:and look at it, there's a lot going on with the shelf that's
Mike Graen:really wrong, but most people will only focus on that red box,
Mike Graen:which is the out of stock. And that's where these guys I think
Mike Graen:really will come into place. Because if you think of a
Mike Graen:retailer like Walmart, that's got several 100,000 items, and
Mike Graen:several hundreds, probably thousands and thousands of
Mike Graen:stores, that's a lot of things that can go wrong. And so we
Mike Graen:think that in store conditions and product recognition can play
Mike Graen:a very important part. And I'm gonna cover these really, really
Mike Graen:quickly because I know we've talked about them before how to
Mike Graen:measure OSA algorithms, but in store audits in store audits
Mike Graen:where somebody's actually going in like a Trax or like a Field
Mike Graen:Agent and taking pictures. It's great that you got the pictures
Mike Graen:and we've actually got Matthew here from Trax, going to talk a
Mike Graen:little bit of what's the behind the scenes sauce that actually
Mike Graen:turns that picture into information. Then we got cases
Mike Graen:like its shelf scanning robots like Woodman's has, they have a
Mike Graen:shelf scanning robot that will go into the store and literally
Mike Graen:scan up and down. But something has to take that information and
Mike Graen:turn that scanning into something that's actionable,
Mike Graen:like what where's the out of stocks, or where's the plugs.
Mike Graen:And then a couple of others ones that I think are appropriate are
Mike Graen:things like Focal Systems and some of the companies that we've
Mike Graen:got, like InView, and Focal Systems and SES-Imagotag who
Mike Graen:have fixed readers, same thing. I've got a fixed camera that's
Mike Graen:taking a picture, something behind the scenes has to take
Mike Graen:that information and go, Okay, what's out of stock, what's
Mike Graen:incorrectly priced, etc. So all three of those I think are very
Mike Graen:relevant to the conversation we're gonna have to have,
Mike Graen:because what we're going to talk about now is, okay, what does
Mike Graen:this data capture mechanism, whether it's a robot or a fixed
Mike Graen:camera, or somebody doing an audit in a store, we got to do
Mike Graen:something with that, and point out what the issues are and
Mike Graen:that's what this group is going to talk to you about going
Mike Graen:forward. So without further ado, let me go ahead and introduce
Mike Graen:this group and we're going to have them unmute, and give
Mike Graen:themselves an introduction. We'll just do it in the order
Mike Graen:that we've got here. We'll give them a few minutes to give them
Mike Graen:an introduction and a little bit about their company. A couple of
Mike Graen:them have sent some slides so we'll I'll advance those when
Mike Graen:you're ready. But the first one is Anand from Infilect, and,
Mike Graen:wow, I believe it's what you know, almost 12 o'clock
Mike Graen:midnight, on a Saturday, you get you get to travel, the farthest
Mike Graen:travel award. Congratulations. Go ahead and unmute yourself and
Mike Graen:tell us about yourself and your company, please.
Anand Subramanian:Great. Thank Thank you, Mike and Matt, for
Anand Subramanian:arranging this podcast. I mean, I've listened to some of these
Anand Subramanian:before, and I'm quite excited to be part of this panel. My name
Anand Subramanian:is Anand Subramanian and I'm the co founder and CEO of Infilect.
Anand Subramanian:Infilect is a retail intelligence company. We
Anand Subramanian:specialize in computer vision and image recognition technology
Anand Subramanian:and help global CPG companies and retailers with in store
Anand Subramanian:retail execution insights. So primarily, we work with visual
Anand Subramanian:data, and then convert the visual data into intelligence
Anand Subramanian:and actions that that can that that can be taken in the stores
Anand Subramanian:so that the reps can improve the install execution. As the slide
Anand Subramanian:says, we are a visual intelligence platform, right for
Anand Subramanian:the worldwide CPG brands and retailers. Mike, just move on to
Anand Subramanian:the next slide, please. Yeah, so just a quick understanding about
Anand Subramanian:the problem that Mike just introduced right OSA, on shelf
Anand Subramanian:availability and in store execution. Pretty much every CPG
Anand Subramanian:supplier would want to understand what exactly is
Anand Subramanian:happening in thousands of retail outlets where their products are
Anand Subramanian:placed. A lot of these CPG suppliers have been using
Anand Subramanian:traditional manual audits, right? I mean, all of us know it
Anand Subramanian:is slow, a lot of human errors can happen and not very
Anand Subramanian:efficient, right. So we work in this space of completely
Anand Subramanian:transforming this audit process into a simple click and go
Anand Subramanian:process and use image recognition technology to help
Anand Subramanian:the CPG suppliers, retailers, and even merchandising companies
Anand Subramanian:to understand and execute in store execution in a better way.
Mike Graen:Yep.
Anand Subramanian:Thank you so much.
Angam Parashar:Got it. Maybe I'm next.
Angam Parashar:Thank you. Thank you, Mike. And thank you, Mike,
Mike Graen:Correct.
Mike Graen:for organizing this this podcast, really appreciate it
Mike Graen:and glad to be a part of it. Good afternoon, good morning,
Mike Graen:good evening, everybody, wherever you are. I am Angam, co
Mike Graen:founder and CEO of Parallel Dots. At Parallel Dots, we are
Mike Graen:building a computer vision platform for CPGs and retailers
Mike Graen:globally. So essentially, a very similar problem statement that
Mike Graen:Anand just talked about. We again work with the global CPG
Mike Graen:brands, and retailers to help them understand what their
Mike Graen:visibility is on the retail shelves, help them minimize that
Mike Graen:out of stock, help them maximize their OSA numbers, their
Mike Graen:compliance numbers in general. We do it by analyzing pictures
Mike Graen:of retail shelves by passing them through our proprietary
Mike Graen:computer vision algorithms through which we can determine
Mike Graen:important retail execution KPIs such as you know, of course on
Mike Graen:shelf availability, share of shelf, planogram compliance,
Mike Graen:pricing compliance, those kinds of things. And we also help our
Mike Graen:clients adopt these solutions at a large scale so that they also
Mike Graen:start getting benefits, you know, such that it either
Mike Graen:increases their sales or decrease the cost, helping them
Mike Graen:improve their, their, their numbers. Right, so we also focus
Mike Graen:a lot on that aspect.
Mike Graen:Awesome. Renish, your next.
Renish Pynadath:Thank you, Mike. Hi, good morning, good
Renish Pynadath:afternoon, good evening, everyone. This is Renish, co
Renish Pynadath:founder for Snap2Insight. We are headquartered in Portland,
Renish Pynadath:Oregon. And again, operating the same space. A retail analytics
Renish Pynadath:provider leveraging AI and computer vision to help CPG
Renish Pynadath:suppliers and retailers and to measure and improve their shelf
Renish Pynadath:execution. We also operate in the space of helping some of the
Renish Pynadath:big suppliers to do space service, which is in some sense
Renish Pynadath:a precursor to space planning, and followed by space retail
Renish Pynadath:execution. So we we today use AI and computer vision to help
Renish Pynadath:these suppliers mostly and sometimes their third party
Renish Pynadath:merchandisers or brokers, or their distributors who go into
Renish Pynadath:stores to better execute against their plan. So yeah, in a
Renish Pynadath:nutshell, that's what we do, again, using computer vision and
Renish Pynadath:image recognition technology. Nice to meet everyone here.
Mike Graen:Awesome. Thank you Renish. Last but not least, Mr.
Mike Graen:Greene. Matt, you want to unmute and tell us about yourself and
Mike Graen:your company.
Matt Greene:Yes, thank you. Thanks, Mike, for organizing
Matt Greene:this, and thanks to Anand, Angam, and Renish, for your
Matt Greene:contributions to what I consider a really, really cool field. I'm
Matt Greene:excited to talk to you guys about our perspective from Trax
Matt Greene:in artificial intelligence and image recognition. This panel, I
Matt Greene:think, is going to be an awesome way for all of us to learn from
Matt Greene:each other and I'm really excited kind of for the next
Matt Greene:steps. Our companies are really aimed at the same main
Matt Greene:objectives, I won't bore you guys and tell you exactly what
Matt Greene:it is that we're out to solve. But in a nutshell, it's
Matt Greene:optimizing your in store field execution using artificial
Matt Greene:intelligence. I've been with Trax for about five years, I
Matt Greene:came to Trax by way of an acquisition in 2018. And prior
Matt Greene:to that worked at a company called Currie, which kind of was
Matt Greene:one of the early pioneers of in store crowdsourcing. Previous to
Matt Greene:that started in CPG, at Nielsen. Learned a lot about what
Matt Greene:customers were looking for in store and that there was a real
Matt Greene:market need in this industry. And as I said, I'm really
Matt Greene:excited to talk to you about what it is that we've learned
Matt Greene:since then. If you wouldn't mind, Mike just advancing next
Matt Greene:slide. So a little bit about our company. We have a very global
Matt Greene:reach, but we like to think it's very local impact. We are in
Matt Greene:over 90 countries worldwide. Our, America's headquarters,
Matt Greene:where I live is in Denver, Colorado. Our R&D headquarters
Matt Greene:is in Tel-Aviv, Israel. And as I mentioned earlier, we have a
Matt Greene:couple of companies that we have acquired over the over the
Matt Greene:years, one of which is where I came from, that really allowed
Matt Greene:us to look at retail execution not just as AI but as kind of an
Matt Greene:end to end service model. And we'll talk a little bit about
Matt Greene:that in a sec. Next slide, Mike.
Mike Graen:Oops, sorry.
Matt Greene:Yeah, so we work with 30 of the top 50 CPGs or
Matt Greene:there about and as I mentioned, kind of partner globally. You
Matt Greene:know, we feel like our global reach and impact is definitely
Matt Greene:something that many CPGs of this side need. Next slide. And this
Matt Greene:just gives you a little bit of understanding of kind of what
Matt Greene:our core technology does. So we have really two main components,
Matt Greene:we do production or I guess image processing of a shelf in
Matt Greene:the cloud. We also have on device recognition, both of
Matt Greene:which are kind of served for the purpose of a certain use case,
Matt Greene:whether it is a customer that is in field that is trying to be
Matt Greene:more accurate and efficient with their day, or whether it's a
Matt Greene:headquarter call that needs more information, more understanding
Matt Greene:of what's going on in the field. We have two kind of primary
Matt Greene:methods of recognizing the SKUs that are on a shelf. Next slide,
Matt Greene:Mike. What happens here and there's there's a bit of a build
Matt Greene:is these images come back, they are then stitched together
Matt Greene:through our engine. And you can see basically simulate what's
Matt Greene:going on in an aisle, allowing anybody from a headquarter
Matt Greene:perspective or a field perspective to see exactly
Matt Greene:what's what's happening in that aisle. And part of the value of
Matt Greene:this is putting together disaggregate images that come in
Matt Greene:from the field and turning it into a real picture. So you
Matt Greene:could virtually see walk and kind of understand what's
Matt Greene:happening and in multiple different, you know, regions,
Matt Greene:cities, countries globally, and it leads to a very cool
Matt Greene:experience. And finally, next slide. And then obviously, one
Matt Greene:of the most important things, if not the most important thing
Matt Greene:about our company is we try to take all of this information and
Matt Greene:turn it into a really translatable asset for you to
Matt Greene:react to. And I think that if you talk to any of the other
Matt Greene:panelists here, they would say the same thing that data is, you
Matt Greene:know, information overload is a real thing right now and if you
Matt Greene:can't translate that to something that matters for the
Matt Greene:customer to make a real decision on, then you're probably not
Matt Greene:serving as much of a purpose as you need to. And last slide,
Matt Greene:Mike. I want to just talk a little bit about where we're
Matt Greene:seeing image recognition going and I'm sure this will come out
Matt Greene:kind of in the further panel discussion. But we'd like to
Matt Greene:think of image recognition at Trax as a backbone, almost a
Matt Greene:commodity of what's going on in store and the value of image
Matt Greene:recognition today, it's very different than the value of what
Matt Greene:it was five years ago, ten years ago, when it was a very new
Matt Greene:early stage technology. So of course, it's the table stakes
Matt Greene:you need to do, you need to annotate what's going on in the
Matt Greene:store. But what really matters is what you're doing to react to
Matt Greene:it. So we're very much invested from a company perspective in
Matt Greene:building an on demand flexible labor source that actually
Matt Greene:executes on the results of image recognition to close holes as
Matt Greene:quickly as you know a customer needs them to be when a shelf is
Matt Greene:not looking like it needs to. And then also, we're investing
Matt Greene:very much in shopper activation. Mike made a comment earlier
Matt Greene:about changing consumer behavior, and how that's a real
Matt Greene:thing. How if a product is not on the shelf, you're almost
Matt Greene:instantly going to lose them to a competitor. And we're very
Matt Greene:much focusing on maintaining consumer relevance through a
Matt Greene:shopper activation field force that again, both of these
Matt Greene:technologies are leveraging image recognition, to maximize
Matt Greene:what's happening in store. And again, that's a little bit about
Matt Greene:Trax, looking forward to the next stage of the discussion.
Mike Graen:Awesome, thank you very much, Matt, appreciate it.
Mike Graen:I'm gonna I'm gonna leave these particular things up here,
Mike Graen:because what you'll see is QR codes for each one of their
Mike Graen:LinkedIn profiles. If you have any questions that you want to
Mike Graen:ask specifically to one of them, then that'll give the folks
Mike Graen:online opportunity to do that. Got several questions here, but
Mike Graen:actually, I think some of you guys have already answered them.
Mike Graen:So I'm going to do some combination, little questions
Mike Graen:here, which is the role of computer vision okay, you've
Mike Graen:showed me what you've done. But what are the real things that
Mike Graen:you're trying to solve? What are you guys solving for? And the
Mike Graen:second kind of follow up question is, who is the
Mike Graen:customer? Is this a retailer that wants to know how they're
Mike Graen:doing? Is it a brand owner or a CPG company that wants to know
Mike Graen:how they're doing? Is it a third party merchandiser? What
Mike Graen:exactly, so what is it doing, and then exactly who your
Mike Graen:customer is. And I'll just, I'll just open it up for all of you
Mike Graen:guys, and just come off mute and try and address that question.
Anand Subramanian:Okay, I can take the first question. I think
Anand Subramanian:computer vision again is a means to generate or digitize the data
Anand Subramanian:that we generate from what from the stores and the most of the
Anand Subramanian:value comes from the data, right. And the data that is
Anand Subramanian:coming out of this algorithms is needs to be analyzed and
Anand Subramanian:insights and actions need to be generated and that is where the
Anand Subramanian:key is. And it is again, just measuring the data is not going
Anand Subramanian:to help anyone, right. I mean, it's probably the first step but
Anand Subramanian:taking the data, taking the insights and actually making
Anand Subramanian:them into actions and then executing them in the store is
Anand Subramanian:what is going to help suppliers or retailers or even the, the
Anand Subramanian:merchandising agencies that you mentioned, to improve their
Anand Subramanian:outcomes, right. So that is where I feel and but but
Anand Subramanian:computer vision is a great tool, it has tremendously improved the
Anand Subramanian:overall process of collecting retail execution data from from
Anand Subramanian:stores, right, because of all the automation and the accuracy
Anand Subramanian:advantages it brings them right. So that is my thought.
Mike Graen:So what I heard you say is it allows you to
Mike Graen:understand what's going on in store. And I heard I think I
Mike Graen:heard you say as both retailers, CPG, brand owners, and
Mike Graen:merchants, merchandisers potentially are customers of
Mike Graen:this. There are probably different ways that data is
Mike Graen:collected that we'll get to in a second but all of them want to
Mike Graen:know what is exactly going on on the retail shelf. Anybody else
Mike Graen:have any builds to that?
Angam Parashar:Yeah, I can add something to it. So yeah, I
Angam Parashar:completely agree with and on this, you know, and what we have
Angam Parashar:seen from our experience is that there's so many different
Angam Parashar:players who can who can benefit from it, right? We have seen
Angam Parashar:well CPGs, one of our first clients to begin with. And of
Angam Parashar:course, there's a clear value prop for them right how they can
Angam Parashar:make their sales reps or their other agent field reps in a more
Angam Parashar:efficient, but we have seen increasing interest from
Angam Parashar:retailers and all format retailers beat from the largest
Angam Parashar:supermarkets, hypermarkets or even convenience store chains
Angam Parashar:all across the globe, being interested in this where their
Angam Parashar:primary objective being they want to reduce out of stock,
Angam Parashar:they want to increase their availability and reduce out of
Angam Parashar:stock. And to also merchandising agencies all across the globe,
Angam Parashar:they have seen, they have also they're also realizing that how
Angam Parashar:they can save their reps by their stores, right? And make
Angam Parashar:the entire piece, entire supply chain much more efficient,
Angam Parashar:right. And everybody benefits from it right. Not just
Angam Parashar:retailer, the CPGs, and agencies, everybody benefits
Angam Parashar:from it. So this has this has been what, and merchandising
Angam Parashar:agencies and retailers are seeing increasing adoption from
Angam Parashar:from these two different layers.
Matt Greene:I would agree with everything you just said. I
Matt Greene:mean, I think we look at it like this is a really, really dynamic
Matt Greene:industry where consumer behavior is changing on the fly. And
Matt Greene:honestly, merchandising behavior is changing on the fly too. If
Matt Greene:you're walking at least in the US, if you're walking around a
Matt Greene:retail store, even during peak shopping hours, it is not a
Matt Greene:strange thing anymore to see five different people and five
Matt Greene:different badges doing five different things at any given
Matt Greene:time, merchandising really in real life. And our goal is to
Matt Greene:make sure that we are informing that merchandising activity on
Matt Greene:behalf of a brand with the most updated relevant in store
Matt Greene:condition information as possible. So that we can put a
Matt Greene:product on shelf for on behalf of that brand or the
Matt Greene:merchandiser can put it can put a product on shelf on behalf of
Matt Greene:that brand with as minimal downtime as possible. Talking a
Matt Greene:little bit about changing consumers, Mike, you made a
Matt Greene:comment at the beginning that is so spot on that if a product is
Matt Greene:not on shelf, the consumer will not be brand loyal, and they
Matt Greene:will pick something else. And anytime that that experience
Matt Greene:happens to your everyday shopper, the brand is basically
Matt Greene:missing an opportunity to keep a customer and there's a decent
Matt Greene:chance that they're going to find something else in real time
Matt Greene:and ultimately make a different decision that could impact their
Matt Greene:their behavior in the in the, you know, future purchasing. So
Matt Greene:we aim to make all of that a more seamless process, make sure
Matt Greene:that all those merchandisers that are always running around
Matt Greene:the store, especially now, have the right information and
Matt Greene:merchandise in a really effective way.
Mike Graen:Great. So
Renish Pynadath:I want add one more thing.
Renish Pynadath:Yeah, go ahead Renish. I was going to call on you next.
Renish Pynadath:Yeah, agree with all the points that I heard here. One thing
Renish Pynadath:that we also started to notice is that, using the power of
Renish Pynadath:computer vision, we are able to see the shelf conditions, but we
Renish Pynadath:also see a role for technology to help the merchandiser or the
Renish Pynadath:field rep to take that next best action on the shelf. How can we
Renish Pynadath:reduce their cognitive load by using technology. Suggesting the
Renish Pynadath:next best alternative in terms of which product is stocked, if
Renish Pynadath:the shelf has an out of stock, right? How to make real time
Renish Pynadath:decisions, right? Without, and without data, it's very, very
Renish Pynadath:difficult, right? And sometimes the folks at the shelf are in
Renish Pynadath:some sense, not that skilled to be able to take those heavy
Renish Pynadath:cognitive load. And they are sometimes working against very
Renish Pynadath:aggressive timelines, right of covering many stores across
Renish Pynadath:several categories. Right. So how do we enable them to take
Renish Pynadath:those right decisions on the shelf? Just like we are helping
Renish Pynadath:a shopper to take those right shopping decisions, right? So we
Renish Pynadath:see that as an addition extension to what we do, in
Renish Pynadath:addition to looking at how the shelf is.
Mike Graen:Okay, thank you. Great answers from all of you.
Mike Graen:I'll put a couple of you on the spot here. Matthew, I'm gonna
Mike Graen:start with you. How are these images collected? Obviously,
Mike Graen:what your model is somebody go in the store, whether it's a
Mike Graen:retailer, or merchandising, or brand or take a picture. We also
Mike Graen:talked about earlier about robotic cameras, or I'm sorry,
Mike Graen:shelf scanning robots in store fixed cameras, etc. What's the
Mike Graen:most predominant way you see today the images being captured?
Mike Graen:And do you think that's going to change over time?
Matt Greene:Yeah, it's a really good question. Thank you. You
Matt Greene:know, I think the most predominant method certainly
Matt Greene:will change. But from my perspective, it is still human
Matt Greene:with a mobile application. And that you know, that human is
Matt Greene:somewhat agnostic. It could be someone from a mobile crowd,
Matt Greene:right? Someone that has an app that you are signed up that is
Matt Greene:using, you know, to go in store, get paid and collect, collect
Matt Greene:data. It could be a merchandiser themselves, it could be someone
Matt Greene:from a traditional broker, like Advantage, like Crossmark, like
Matt Greene:Acosta, like any of the above, or it could be an actual sales
Matt Greene:rep. And so that is certainly from Trax's perspective where
Matt Greene:most of our business is, is with a human using a mobile
Matt Greene:application. But the human might kind of look and feel very
Matt Greene:different based off of based off of you know, who they work for,
Matt Greene:and how they work. Now, going forward, we also have fixed
Matt Greene:cameras, that has been part of our company strategy for some
Matt Greene:time. And the holy grail is to basically have this action done
Matt Greene:through, you know, through not having humans. Whether that's a
Matt Greene:drone, whether that's a camera, whether that's a dog that walks
Matt Greene:down the aisle with, you know, a phone attached to them. There's
Matt Greene:there's a lot of need for automation and continued
Matt Greene:automation in space to make sure that the cost of data collection
Matt Greene:is as low as it needs to be for a technology like this to really
Matt Greene:scale and make sense for our customers. Does that answer your
Matt Greene:question, Mike?
Mike Graen:100%. Yeah, because you know, labor, whether it's
Mike Graen:some merchandiser, or brand owner or retailer, you really
Mike Graen:want those folks actually taking action on the output of your
Mike Graen:solution and not being a data collection device, if at all
Mike Graen:possible. So I believe, I don't know about the drone thing. I
Mike Graen:don't know if I want to get buzzed in the store with a drone
Mike Graen:taking pictures, but I get your point. And I've never seen a dog
Mike Graen:with a camera on it, that could happen, I suppose. But to me,
Mike Graen:certainly seeing shelf scanning robots that are very prevalent
Mike Graen:on the industry, starting to see fixed cameras, starting to see
Mike Graen:more and more of that technology. And then okay, I got
Mike Graen:all these pictures now what do I do with them, which is was part
Mike Graen:of your pieces. Renish, question for you. So how does this
Mike Graen:processing occur? Is it like, feed, feed all these pictures
Mike Graen:into something and you guys crank on it for over the
Mike Graen:weekend, and on Monday morning, you give me the results. So you
Mike Graen:know what is, what is actually the processing behind the scenes
Mike Graen:look like in this thing?
Renish Pynadath:Yeah. Thanks, Mike. Again, let me take this in
Renish Pynadath:two parts, right, like, where does processing happen, right,
Renish Pynadath:and today, I think like others mentioned, like these pictures
Renish Pynadath:are being captured by different types of endpoints, whether it's
Renish Pynadath:a device, whether it's reps phone, whether it's shelf edge
Renish Pynadath:camera or shelf scanning. And with the recent advances in this
Renish Pynadath:in the compute capability of these devices, what we are
Renish Pynadath:seeing is, technology is able to do a lot of these, the heavy
Renish Pynadath:lifting on the device, which helps us avoid network transfer
Renish Pynadath:cost and giving faster results, right. So we see that the
Renish Pynadath:processing itself is moving more and more closer to where the
Renish Pynadath:action is happening. And when you look at how does it happen,
Renish Pynadath:these images are being processed by computer vision technology,
Renish Pynadath:where you're trying to extract features of what we see in those
Renish Pynadath:images and then compare against what is expected, right. And
Renish Pynadath:that helps give actionable information whether it is
Renish Pynadath:comparing against a distribution list expected or whether it's a
Renish Pynadath:planogram, where, for example, if you have prospects filed for
Renish Pynadath:a particular shelf, you can compare against that. In certain
Renish Pynadath:cases for a supplier you're trying to measure against their
Renish Pynadath:contracts for a contract compliance. A supplier has a
Renish Pynadath:particular contract with a particular chain, looking for
Renish Pynadath:compliance against that. So there are different aspects of
Renish Pynadath:shelf execution that is being compared against with the data
Renish Pynadath:that was extracted from the majors. So that's, that's the
Renish Pynadath:processing part, right.
Mike Graen:Got it. So let me let me ask you guys, I want to
Mike Graen:go back to a slide that I shared before. It's this one here that
Mike Graen:basically kind of lays out the different challenges and there's
Mike Graen:probably more I mean, just to be clear, there's probably more,
Mike Graen:but as you look at this thing, what kind of data and images and
Mike Graen:information do you need to basically tell a retailer or
Mike Graen:brand owner I see a label, I see no product, there's obviously an
Mike Graen:on shelf availability issue. What data do you practically
Mike Graen:need to be able to do that? I'll start with you Anand. What data
Mike Graen:do we need for your algorithm to be able to tell us that?
Anand Subramanian:Yeah, I mean, I think the basic information is
Anand Subramanian:the master data from the brand or, or the product category,
Anand Subramanian:right? So what are the products that need to be recognized
Anand Subramanian:because AI is in a state where it needs to be still taught
Anand Subramanian:right to understand what product is what so there is some kind of
Anand Subramanian:data that is required. Of course, images from the shelf,
Anand Subramanian:which is basically the images that we capture from the stores
Anand Subramanian:with just these two set of data, algorithms can be customized,
Anand Subramanian:and they can be as accurate as, as ground truth, right? Like,
Anand Subramanian:like a human would see it. These are the two set of information.
Anand Subramanian:And as we go, work with more and more customers, we see that the
Anand Subramanian:amount of data that is needed from the customers is, is not a
Anand Subramanian:lot, right, a set of pictures from the stores, as well as the
Anand Subramanian:master data from the customer, is typically what is needed to
Anand Subramanian:customize algorithms which are very accurate. And that can
Anand Subramanian:digitize these visual images from the stores and then
Anand Subramanian:generate data out of it.
Mike Graen:So you need a picture of the shelf, and you
Mike Graen:need a really good quality product database that you can
Mike Graen:match it up against and go yeah, that's a problem. That what I'm
Mike Graen:hear you saying? Alright, so if we expand that, beyond this, and
Mike Graen:I'm going to put Matt on the spot now. Pricing, this one is
Mike Graen:very important. There are some places in places like New York
Mike Graen:and New Jersey, if there is not a match of compliance, retailers
Mike Graen:can get fined a lot of money because the price at the
Mike Graen:register and the price at the shelf doesn't match. So in
Mike Graen:addition to what what what Anand just mentioned, what else do you
Mike Graen:need to be able to be able to say there's a pricing discrepancy?
Matt Greene:Yeah. So pricing is is probably one of the most, I
Matt Greene:don't know if the group agrees with me here. But from my
Matt Greene:perspective, one of the most misunderstood components of
Matt Greene:image recognition. OCR technology has been around for a
Matt Greene:long time, and one would think that you could use OCR
Matt Greene:technology to basically validate or invalidate a price fairly
Matt Greene:easily, and then correlate that back to what SKU is available or
Matt Greene:is not. And that's, that's true to an extent. The reality is,
Matt Greene:though, is that especially kind of because we're working with
Matt Greene:customers that are in a lot of different places of the world,
Matt Greene:and if you really look at pricing at some of the smaller
Matt Greene:stores like bodegas, let's use your example of New York City.
Matt Greene:There's a lot of handwritten prices, there's a lot of things
Matt Greene:that are not easy to recognize, even for the human eye. So I'll
Matt Greene:start by just saying that pricing, while the technology is
Matt Greene:there, the real kind of special sauce is using the price,
Matt Greene:getting it at scale, and then correlating it to the product
Matt Greene:within the photo that it comes back. That really is a hard
Matt Greene:thing to master, and it's something that when you start
Matt Greene:looking at a lot of images, it becomes very real and apparent
Matt Greene:that pricing is not as simple as you might think. So that's the
Matt Greene:first point. The second point that I would say is that we we
Matt Greene:use pixelation to basically determine where the price is and
Matt Greene:where the product should be to then create almost a an
Matt Greene:artificial link to what that hole actually is. And that's how
Matt Greene:we would tell a customer what the hole on the shelf is, is
Matt Greene:through basically correlating hole with price and then how
Matt Greene:close in proximity it is. And then obviously, there are
Matt Greene:processes around you know, quality assurance and cleansing
Matt Greene:that makes sure that you know, that's working effectively. But
Matt Greene:to answer your question, Mike, price is an enormously impactful
Matt Greene:thing. And it tells a real story when things are priced correctly
Matt Greene:or incorrectly that you know becomes a lot of value to a
Matt Greene:brand and then also a retailer to see at scale.
Mike Graen:Got it. Perfect. Anand, I'm gonna put you on the
Mike Graen:spot, plugged item. Plugged item basically says there is a label,
Mike Graen:there is a product, but your tool is going to tell me that
Mike Graen:that product is not the right product for the shelf. And that
Mike Graen:causes two problems. Number one, the products not on the shelf.
Mike Graen:Number two, when the person comes to stock that product
Mike Graen:cause it finally shows up, they go to stock and they can't find
Mike Graen:out where it supposed to go because somebody put something
Mike Graen:else in the way. So, what data do you need to be able to say
Mike Graen:that item is a plugged or an incorrect item or sometimes they
Mike Graen:call it they've expanded the facings of it. What do you need
Mike Graen:to be able to accurately tell that to a retailer or a CPG
Mike Graen:company?
Anand Subramanian:Yeah, so that the image of the shelf from the
Anand Subramanian:image of the shelf itself the pricing stickers can be decoded
Anand Subramanian:like electrical. How Matt was mentioning using OCR technology.
Anand Subramanian:Of course there are challenges because these stickers are very
Anand Subramanian:small and handwritten sometimes in some of the stores these are
Anand Subramanian:not very in standard format, right. But I think the two
Anand Subramanian:inputs that are needed is the record recognizing the sticker,
Anand Subramanian:as well as recognizing the product that is plugged. Right.
Anand Subramanian:So and then match them together because the the sticker usually
Anand Subramanian:has the SKU name as well as the price, right? And and the
Anand Subramanian:product recognition will tell what is the exact product that
Anand Subramanian:is placed there. And using both of these information, we can
Anand Subramanian:identify whether it's a plugged item or not, right. And these
Anand Subramanian:are the two sets of information. And again, with with computer
Anand Subramanian:vision, just by scanning the shelf, these two informations
Anand Subramanian:are extracted automatically with algorithms and then compare it
Anand Subramanian:and then these alerts can be generated.
Mike Graen:Got it. Perfect. All right. Angam on the last one is
Mike Graen:probably the hardest one. So I figured I'd leave the last one
Mike Graen:to you. So these yellow boxes, basically say the product is not
Mike Graen:there nor is the label there. So there is supposed to be a label
Mike Graen:there that is not there, it got skipped. In addition to the
Mike Graen:picture of the shelf, what do you need from the retailer to be
Mike Graen:able to say there is a distribution void, or there's a
Mike Graen:missing label that should be there that's not?
Angam Parashar:No, I think there's not other input required
Angam Parashar:from the retailer, we need to the CV, computer vision
Angam Parashar:technology needs to be smart enough to identify gaps in the
Angam Parashar:images, right? To understand that, okay, there is supposed to
Angam Parashar:be a product here, there is supposed to be an SKU here, but
Angam Parashar:there's a big gap, the particular red one that you see
Angam Parashar:right out of stock box that you see it's the CV technology needs
Angam Parashar:to be smart enough to identify that gap. And also see that
Angam Parashar:there is no price label we need that right. The first step when
Angam Parashar:you're doing pricing is to segment each and every price tag
Angam Parashar:written right? Once you segment it, then you read it. So if
Angam Parashar:there is nothing to segment, you know that there is no price tag,
Angam Parashar:and ultimately, you tell that, okay, this is just an empty
Angam Parashar:space, there's supposed to be something and nobody knows what
Angam Parashar:was supposed to be there, right? So you can go to the planogram
Angam Parashar:image and try to figure out what could be here. But otherwise,
Angam Parashar:you know, you have to detect the empty space and the missing
Angam Parashar:label that there is no missing label to come up with this
Angam Parashar:information.
Mike Graen:All right, so I'm gonna, I'm gonna put you guys on
Mike Graen:the spot a little bit, and then just just to let the audience
Mike Graen:know, this will be my last kind of question before I kind of
Mike Graen:open it up for any questions you guys might have on this kind of
Mike Graen:section. Every one of you have been mentioned product database,
Mike Graen:in one way, shape, or form. You gotta have a product database
Mike Graen:that you're matching it to. Talk to us about that because my
Mike Graen:sense is each one of you are probably using your own product
Mike Graen:database, you're expecting it from the retail or the CPG
Mike Graen:owners. I would imagine that this is probably one of the
Mike Graen:biggest challenges in your space, which is, brands spend
Mike Graen:millions of dollars a year just changing packaging to make it
Mike Graen:look like new and improved, right. So how in the world, and
Mike Graen:you certainly get products, while this particular example is
Mike Graen:very clean, you'll get things turned to the side, you'll get
Mike Graen:things knocked down, you get all those other kinds of issues that
Mike Graen:you have to deal with. Talk to us about the importance of the
Mike Graen:product image library to be able to do this kind of work. And you
Mike Graen:know, what's the roadmap in the future look like for product
Mike Graen:product databases?
Angam Parashar:Me, I can go first.
Mike Graen:Sure go ahead.
Angam Parashar:Yeah, so this is, as you rightly said, this is
Angam Parashar:a huge problem, right? To maintain the product library,
Angam Parashar:and almost to a certain extent some moving goalposts, right.
Angam Parashar:It's, you're never going to achieve that. Right? Every time
Angam Parashar:there's a new SKU, you can achieve it for a particular
Angam Parashar:region, the button, then you move to another market. And then
Angam Parashar:again, there are newer SKUs, right. So it's a it's a big
Angam Parashar:challenge. I think this to a certain extent, can be solved
Angam Parashar:using AI technology, how strong your AI technology is,
Angam Parashar:particularly how much data do you require when you want to
Angam Parashar:train your AI? Right? If you require thousands and thousands
Angam Parashar:of shelf photos, it's going to be a nightmare, right? If you
Angam Parashar:require only few tens of shelf photos, the problem becomes much
Angam Parashar:more manageable. But what you're always trying to do, you're
Angam Parashar:trying to manage the problem, right, unless somebody can form
Angam Parashar:a universal database, which can be applicable everywhere.
Angam Parashar:Another thing to do this, and this is something that we have
Angam Parashar:been doing with other clients as well, is to have very deep
Angam Parashar:integrations with the clients, right? So the moment there's a
Angam Parashar:new SKU, or a new packaging of a particular SKU, right, you
Angam Parashar:quickly get it within your database and you can start
Angam Parashar:training it right. So this is something which has been helping
Angam Parashar:us where we have really integrated our technology with
Angam Parashar:their, with their databases to ensure that we are we always
Angam Parashar:remain aware of what's the latest SKU is and ultimately
Angam Parashar:what you also need to do is really reduce your AI training
Angam Parashar:time so you cannot take weeks to train your AI because by then
Angam Parashar:the packaging will change again right. So you have to be on your
Angam Parashar:toes. It should happen within a matter of a few days, if not
Angam Parashar:hours, right and and then you know, you should start you
Angam Parashar:should start delivering a very high level of accuracy right on
Angam Parashar:day one because you know that's when the value, that's when the
Angam Parashar:client will get the value. Right?
Mike Graen:Perfect.
Anand Subramanian:Yeah, one point adding to what Angam
Anand Subramanian:mentioned, new product launches as well as custom design changes
Anand Subramanian:are definitely happening. And they happen at different rates
Anand Subramanian:with different customers and also depends on the channel as
Anand Subramanian:well, right. So integrating into the customer database or
Anand Subramanian:understanding the launch calendars can give a heads up,
Anand Subramanian:right. So that is one of the key ways we also try to understand
Anand Subramanian:what is going to come right. And usually brands have a launch
Anand Subramanian:calendar, like what are the new SKU is or is there going to be a
Anand Subramanian:combo pack or some kind of an offer? Right, all these are
Anand Subramanian:planned by the point or by the by the suppliers, right? Or even
Anand Subramanian:by the retailers. So with that, and then very, very
Anand Subramanian:sophisticated algorithms, then these can be handled in a better
Anand Subramanian:way. Right. Another approach is, there are I mean, algorithms are
Anand Subramanian:now smart enough to identify some of the new products that
Anand Subramanian:are appearing on the shelf, which they have not been
Anand Subramanian:trained, not trained for. Right. So let's say there are some 10
Anand Subramanian:SKUs of head and shoulder here, and then the suddenly we see 11
Anand Subramanian:to 12 SKU, algorithms can pick and say that this is something
Anand Subramanian:new, right? So so we sometimes go proactively to the customer
Anand Subramanian:and say, Hey, this is something that is these are some couple of
Anand Subramanian:escapes that we see new in the the field, do you do you want to
Anand Subramanian:add them to the recognition, right? And then they come back
Anand Subramanian:and say, yeah, these are some new launches. And then yes, of
Anand Subramanian:course, right, and then then we start tracking them. So there
Anand Subramanian:are multiple ways in which new product launches, as well as
Anand Subramanian:design changes can be handled. And definitely, the technology
Anand Subramanian:is very sophisticated now to handle these pretty much with a
Anand Subramanian:very, very fast response time.
Matt Greene:I would agree with everything that's been said.
Matt Greene:This is a huge, I don't really want to call it a pain point,
Matt Greene:but I think it's something that if an image recognition CV
Matt Greene:solution does not take it super seriously, then they'll fall
Matt Greene:behind against technology and others that are one of the most
Matt Greene:valid points about why having a updated image library and master
Matt Greene:database is so important is because when you have the right
Matt Greene:information at a SKU by SKU level, and you can recognize in
Matt Greene:the photo exactly what that SKU is, you can then tie back very
Matt Greene:interesting custom attributes to basically what will build the
Matt Greene:end outcome or KPI for a customer. And in order to do
Matt Greene:that, you have to tag it at a SKU level. In order to tag it
Matt Greene:correctly at a SKU level, you need to have the right image to
Matt Greene:train on. So exactly like what these guys said, I think the
Matt Greene:solve right now is through creating you know more
Matt Greene:autonomous processes within our individual companies that will
Matt Greene:allow for, you know, a certain queue to be created of things
Matt Greene:that need to be looked at and retrained as possibly a design
Matt Greene:variant, and acting on that as quick as you humanly can. The
Matt Greene:long term, you know, big utopia for this industry is an updated
Matt Greene:relevant master syndicated database that you know, everyone
Matt Greene:can look at as the source of truth, and that just doesn't
Matt Greene:exist right now. So
Mike Graen:Yeah, great point. I mean, it feels a little bit like
Mike Graen:you guys are almost differentiating yourself based
Mike Graen:upon the quality of your product recognition library, which is
Mike Graen:great, I suppose. But honestly, that to me is not the secret
Mike Graen:sauce, I'd much rather you'd be focusing on different ways to
Mike Graen:look at the same problem and solve different problems. I
Mike Graen:wonder if there was some kind of industry and industry standard
Mike Graen:would be better. I'm gonna open it up to a guy by the name of
Mike Graen:Mike Price, who always has really tough questions. This
Mike Graen:Mike is easy on you guys, but Mike price, I'll give it to you.
Mike Graen:He's gonna give it to you. And he's also running on European
Mike Graen:time so he's probably a couple couple of hours away from
Mike Graen:sleeping too. So Mike, I'm gonna open it up to you. Hopefully,
Mike Graen:you'll be able to talk if I've given you the permission to
Mike Graen:talk. So go ahead and unmute and ask your question for us.
Mike Price:Thanks, Mike. I'm sure I'm surprised you haven't
Mike Price:banned me from your sessions with the questions that I've
Mike Price:been asked over the
Mike Graen:I keep I keep saying disinvite but it doesn't do it.
Mike Graen:It just comes back and gives me more questions. So you're ok,
Mike Graen:you always ask great questions, Mike.
Mike Price:Well, I'll speak to the guys that stay in front
Mike Price:because they've managed to disable me from all their
Mike Price:webinars. Anyway, moving on. Moving on. I never I'm never I
Mike Price:never get knocked off of this. Anyway, moving on. So my
Mike Price:question to you guys is, I've been involved with this or was
Mike Price:involved with this when I was at Unilever since 2014. There's a
Mike Price:lot of you in the space now I keep a list. And I've got 43
Mike Price:partners, as I would call you on my list across mobile, fixed
Mike Price:cameras and robotic solutions. So my question is, how the hell
Mike Price:do you stand out and what makes you different? What is your USP
Mike Price:because most CPGs go through a pitch of five or six potential
Mike Price:partners, and it's really difficult. So that's my
Mike Price:question. Thanks, Mike.
Mike Graen:Yep, thank you, Mike. I told you I was gonna ask
Mike Graen:you tough questions. Please do me a favor and don't
Mike Graen:differentiate yourself via price because I don't want to get into
Mike Graen:antitrust things. What is your what is your point of
Mike Graen:differentiation? I think that's a very fair question. We'll
Mike Graen:start, Renish, why don't you go ahead and kick it off.
Renish Pynadath:Yeah, so I think, I think like, like we all
Renish Pynadath:agree here, the ability to kind of get the training right in a
Renish Pynadath:very cost effective ways, one of the biggest challenges in this
Renish Pynadath:industry, right. So we talked about master data, about
Renish Pynadath:products, master majors, right? So we differentiate ourselves
Renish Pynadath:being being able to train up data through both stock images
Renish Pynadath:that we, for example, what we see on Amazon product listing,
Renish Pynadath:and real world shelf images, right. So it's a mix of both
Renish Pynadath:stock images and real world images, that helps us to be very
Renish Pynadath:up to date when it comes to being able to recognize products
Renish Pynadath:through our models. And that also gives you to actually
Renish Pynadath:differentiate yourselves when it comes to cost that you can offer
Renish Pynadath:because the moment you can train up without too much of real
Renish Pynadath:world images, which are costly to acquire, especially the long
Renish Pynadath:tail part of it, like I think Angam was talking about it, you
Renish Pynadath:go to any new market, any new part of the country, you are
Renish Pynadath:going to get new products in the same category that you're
Renish Pynadath:operating at. How do you minimize that and be able to
Renish Pynadath:give a very cost effective solution. So that is one right?
Renish Pynadath:Where we have built our platform to be able to use stock images,
Renish Pynadath:along with real shelf images to make training faster, and cost
Renish Pynadath:effective, right? The second is, how fast can your system do do
Renish Pynadath:the recognition and then suggest the next best action, right? So
Renish Pynadath:we, we again, differentiate ourselves on speed, right? For
Renish Pynadath:some of the largest snack manufacturer in the world, we
Renish Pynadath:audit about 300,000 stores, and a million fixtures are there and
Renish Pynadath:today, we are able to get the results back to them in under 30
Renish Pynadath:seconds for every fixture that they audit, right. So speed,
Renish Pynadath:again, speed to speed to results, right? That is the
Renish Pynadath:second part that where we differentiate. So those are, I
Renish Pynadath:would say two pillars on which we differentiate in terms of
Renish Pynadath:being fast to be able to train up and cost effective. And the
Renish Pynadath:second is being very fast at the shelf to be able to give that
Renish Pynadath:recommended next best action to the rep who's standing in front
Renish Pynadath:of the shelf to fix the issue of the shelf.
Mike Graen:Awesome. Anand, you want to go next? What
Mike Graen:differentiates you?
Anand Subramanian:Yeah, well I would have told the same thing
Anand Subramanian:that what Renish would've said. I would I would assume that what
Anand Subramanian:Matt, as well as Angam was also going to say the same thing. So
Anand Subramanian:I think the key key here is, of course, think there are
Anand Subramanian:different levers in which we try to stand out right, for example,
Anand Subramanian:speed at which we can customize algorithms and then get things
Anand Subramanian:rolling, right. As well as the next best action part, and there
Anand Subramanian:are a lot of these key things that are there. But I think the
Anand Subramanian:differentiating factor will be to what degree we are able to
Anand Subramanian:do, right? And I think that is where. I mean if you ask feature
Anand Subramanian:by feature, I think anybody who any one of us can build a
Anand Subramanian:feature that is not there with any one of us, right? It's not a
Anand Subramanian:major thing, but at what quality and what precision you are
Anand Subramanian:giving it right. And then I think that is one. The other
Anand Subramanian:interesting part I would add, and maybe others would also say
Anand Subramanian:that that's the differentiation is on the operations part. It's
Anand Subramanian:not just the image recognition capability alone, right? How do
Anand Subramanian:we operationalize this? How do we scale this? And then how do
Anand Subramanian:we make it into a product that the CPG stakeholders are able to
Anand Subramanian:ingest into their regular systems, right? Because unless
Anand Subramanian:we do this in a in a very seamless way, it becomes hard
Anand Subramanian:for them to ingest the data, and then it just becomes a data
Anand Subramanian:overload, right? How are we going to help them to ingest
Anand Subramanian:this into their day to day operations, and then help them
Anand Subramanian:take quick action. So these are some of the key things that we
Anand Subramanian:focus on because we see that of course, image recognition, like
Anand Subramanian:it was in the beginning is it's just a means to an end, right.
Anand Subramanian:But that helps us to get things in a faster, accurate way. But
Anand Subramanian:on top of that, definitely there are a lot of other aspects in
Anand Subramanian:terms of operations, in terms of the analytics capabilities and
Anand Subramanian:other things that differentiates us. Yeah.
Mike Graen:Awesome. Great perspective, great perspective,
Mike Graen:Angam, Parallel Dots, talk to us. What's the differentiator
Mike Graen:for you?
Angam Parashar:Yeah, and you know, just to, as Mike Price
Angam Parashar:said, right, there are 43 companies in his radar building
Angam Parashar:the same thing and the market is crowded, right? So technology
Angam Parashar:too can only be differentiated to a certain extent, right? And
Angam Parashar:for the customers today, it's incredibly hard to differentiate
Angam Parashar:on the basis of technology for them every vendor is essentially
Angam Parashar:the same tech right? So while of course, we focus a lot on the
Angam Parashar:things that you know, Anand said and Renish said and on the tech
Angam Parashar:side, but to build on on what Anand said further right.
Angam Parashar:There's more to it than just a technology piece, right? We have
Angam Parashar:seen a lot of our clients really struggle to piece all these
Angam Parashar:things together when it comes to retail execution there different
Angam Parashar:technologies here, right? You're talking about computer vision,
Angam Parashar:you're talking about, you know, their field operations, you're
Angam Parashar:talking about their ethos data, you're talking about so many
Angam Parashar:different pieces here. Ultimately, everything should
Angam Parashar:combine together and result in something which helps the client
Angam Parashar:increase their sales or reduce the cost, right. It all boils
Angam Parashar:down to two things. And that's where, you know, our approach
Angam Parashar:has been, we have been trying to understand the requirement of
Angam Parashar:our client in a little better way, and trying to forge our
Angam Parashar:approach in a way which helps them get this benefit, right.
Angam Parashar:It's rather than just selling technology, now, we are trying
Angam Parashar:to sell sort of this, our our understanding as well, that we
Angam Parashar:understand the space, and we're gonna help you make these
Angam Parashar:changes in the way you do retail execution, and image recognition
Angam Parashar:is a part of it. So it's more of a consultative approach now that
Angam Parashar:we are effectively choosing to take that to our customers. So
Angam Parashar:that I think is turning out to be a small differentiator in
Angam Parashar:this crowded space for now.
Mike Graen:Gotcha. All right. Last but not least, I know what
Mike Graen:Matt's gonna say, I think. Matt, what's your differentiator?
Matt Greene:Yeah, so I guess everybody is spot on with what
Matt Greene:they're saying. And I think the question, Mike was a really,
Matt Greene:really good one. You know, we know from from a Trax
Matt Greene:perspective, we know and accept that IR image recognition,
Matt Greene:computer vision, is a known commodity, and is very much
Matt Greene:advanced in the last 10 or 12 years. It wasn't always this
Matt Greene:way, but it certainly is now. And I mean, all you have to do
Matt Greene:is talk about the 43 competitors to tell you that it's clearly
Matt Greene:there. And there's there's a very saturated market. So what
Matt Greene:we from a company perspective, have aimed to do is to not think
Matt Greene:of ourselves just as a computer vision provider, but as an end
Matt Greene:to end tech enabled solution that drives and aims to drive an
Matt Greene:enormous ROI for our customers using our service. And what does
Matt Greene:that mean? It means that we don't want to just tell you
Matt Greene:what's on the shelf or not on the shelf. And we don't even
Matt Greene:want to just tell you what you should do based on what's on the
Matt Greene:shelf, we want to make sure that you have the ability to leverage
Matt Greene:a one stop shop to understand what's going on in the market,
Matt Greene:in the stores on the shelf, and then also close the gaps as you
Matt Greene:need to. Merchandising as a whole is extremely ROI driven
Matt Greene:and any change to a service or a service model with merchandisers
Matt Greene:needs to come with a very accelerated return. So said
Matt Greene:differently, if there is someone that wants to change their in
Matt Greene:store execution practice, they need to prove to their own
Matt Greene:company that they're going to drive an increase in sales lift.
Matt Greene:And we believe that our ability to recognize a gap on the shelf,
Matt Greene:and then close it in a very cost, efficient, expedited
Matt Greene:manner with our labor resources and merchandising resources is
Matt Greene:our true differentiator. We also I would say, have a very global
Matt Greene:presence. So it's a pretty compelling conversation for
Matt Greene:someone at a brand's headquarters to look across 50
Matt Greene:different markets at once and measure kind of metro size how
Matt Greene:their perfect store looks in execution and at scale. And
Matt Greene:that's something that, you know, we've seen a lot of success with
Matt Greene:some of our biggest customers where they can basically point
Matt Greene:and click across different geographies and tell a very
Matt Greene:different story than than they have before. And then, you know,
Matt Greene:last but not least, we do have a couple of very tailored packages
Matt Greene:based off of our 12, 12 and a half, 13 years in the industry,
Matt Greene:where we know we've taken a lot of best practices from customers
Matt Greene:that keep telling us the same thing in different ways about
Matt Greene:what it is that they want to see in store, and how they can
Matt Greene:execute differently or what they need, and we've built packages
Matt Greene:and solutions around that. Where we think it's kind of leading
Matt Greene:with industry expertise, because a lot of people are saying the
Matt Greene:same thing. So
Mike Graen:Great, awesome. Y'all did well, and you did it
Mike Graen:without fighting. That's awesome. That's exactly what
Mike Graen:we're hoping for. Appreciate that. I've got
Matt Greene:Great panelists.
Mike Graen:Yeah, I got two more questions. And I'm going to open
Mike Graen:it up for any other questions that people have, because we've
Mike Graen:got about seven minutes left. The first question is kind of a
Mike Graen:two parter. There's a lot we've had a lot of podcasts on here
Mike Graen:from people like crowdsourcing last time, we had Trax and we
Mike Graen:had Field Agent. We had companies that are responsible
Mike Graen:for fixed cameras like Focal, SES-Imagotag and Invue, and we
Mike Graen:have shelf scanning robot companies on as well Badger
Mike Graen:Simbe, Brain, Zippity, just to name a few. If you are going to
Mike Graen:be talking to these solution providers, so no longer about
Mike Graen:retailers and suppliers, but solution providers. Why should
Mike Graen:they come to you and talk to you about your product recognition
Mike Graen:solution rather than building their own because what I've seen
Mike Graen:most of the time, is they build their own. What's the advantage
Mike Graen:to and again, this is for all of you. What's the advantage to
Mike Graen:using you guys versus building their own proprietary image
Mike Graen:recognition software, whoever wants take that one first.
Anand Subramanian:Yeah, I will take it. So when we do see these
Anand Subramanian:solution providers as partners, we have in fact, partnered with
Anand Subramanian:some of them, especially the crowdsourcing companies,
Anand Subramanian:definitely crowds I mean, all of them bring they bring different
Anand Subramanian:perspectives, different strengths, right. I mean, from
Anand Subramanian:from a core computer vision perspective, again, a focus for
Anand Subramanian:our company has been on fine tuning this technology to a
Anand Subramanian:very, very sharp way, that way, image recognition and the data
Anand Subramanian:digitization is done in a very accurate way, right. So, we have
Anand Subramanian:seen, from a technology perspective, our algorithms are
Anand Subramanian:a lot more superior, at the same time, the other players are
Anand Subramanian:bringing in other strengths, right, for example, crowd
Anand Subramanian:sourcing company brings in the field force a cheaper ways of
Anand Subramanian:the conflict case, right. And that is extremely important.
Anand Subramanian:And, and if we are partnering with them, and then it the
Anand Subramanian:overall solution becomes a lot more comprehensive, and then
Anand Subramanian:valuable, right. So that way, we have seen, we do see these as
Anand Subramanian:definitely complementary partners, and an approach where
Anand Subramanian:a combination of strengths work together, and then you better
Anand Subramanian:value to the end customer is, is what what we have seen and
Anand Subramanian:definitely, that's an approach that we're taking.
Mike Graen:Yeah. Awesome. Anybody else?
Matt Greene:Our company vision is very holistic, and not
Matt Greene:dissimilar to what was just said, you know, image
Matt Greene:recognition is one really key piece to that and it's kind of
Matt Greene:the backbone of everything. But if you were to look back on our
Matt Greene:company acquisition strategy, we have acquired and partnered with
Matt Greene:many companies, both from crowdsourcing to robots to
Matt Greene:shopper activation. And I think there's a certain culture within
Matt Greene:our company that acknowledges and knows that there are certain
Matt Greene:players that are always going to be better than us in certain
Matt Greene:things. And our goal is to be the one stop shop for retail,
Matt Greene:not just for image recognition. And that lends itself very well
Matt Greene:to partnership opportunities and overall kind of growth within
Mike Graen:That's great. Thanks. Last question. What's
Mike Graen:the industry.
Mike Graen:the roadmap look like? Where are we going to be two to five years
Mike Graen:from now, I'm gonna say in store recognition, I'm not gonna say
Mike Graen:product anymore because I think there's other opportunities
Mike Graen:besides just product. What's the next two to five years look like
Mike Graen:in terms of in store, you know, recognition of what's going on
Mike Graen:at retail? Each one of you got about a minute or so to close
Mike Graen:it, and then we'll open it up for any final questions.
Angam Parashar:Maybe I can go here and share my thoughts. I
Angam Parashar:think next five years are going to be crazy and super exciting
Angam Parashar:as far as the technology is going to concern right? With the
Angam Parashar:amount of innovation and we can always look at what the past
Angam Parashar:five years have been in the trajectory that we have been on
Angam Parashar:in the past five years, right? Not just we have, the underlying
Angam Parashar:technology has become far stronger. We have now large
Angam Parashar:scale case studies dealing with helping thousands of SKUs at
Angam Parashar:scale, doing analysis on in, in within seconds, right, detecting
Angam Parashar:them in seconds. All of this was probably not possible five years
Angam Parashar:back. And what I mean by this is that five years later, I think
Angam Parashar:that kind of things that we're talking about are only going to
Angam Parashar:get more bizarre in terms of the final output. So for example,
Angam Parashar:you know, right now, we used to talk about image recognition
Angam Parashar:through photos, and we have done it on device, right. I think the
Angam Parashar:next step is going to be videos, right? Where we can do, we'd be
Angam Parashar:processing videos, possibly videos on device as well, with
Angam Parashar:very, very high accuracy. Maybe more KPIs on device itself, you
Angam Parashar:know, some of the more difficult KPIs can also be done on device.
Angam Parashar:All I also think that, you know, a different pieces of
Angam Parashar:technology, as Anand said, would come together in the next five
Angam Parashar:years, right. You will see a more holistic solution either
Angam Parashar:beat by one vendor or beat by one supplier, or multiple
Angam Parashar:suppliers coming in together, and sort of attacking this
Angam Parashar:problem end to end, resulting in a much see more seamless
Angam Parashar:solution than what we have today, right for the for the end
Angam Parashar:clients. So I think those are the few things which are
Angam Parashar:definitely happening in the industry and definitely, you
Angam Parashar:know, something which via suppliers need to be need to
Angam Parashar:know as well as the clients need to know what's what's happening
Angam Parashar:next in the industry.
Mike Graen:Great. Anybody else want to take a shot?
Anand Subramanian:Yeah, I'll add one more point. I think this
Anand Subramanian:is primarily around image recognition in very low light
Anand Subramanian:and very complex scenarios. So a lot of the recognition a lot of
Anand Subramanian:companies have worked in and developed market where there is
Anand Subramanian:nice arrangement of shelf, larger stores, good lighting,
Anand Subramanian:right. But I mean, as all of us know, the complex traditional
Anand Subramanian:trade and the developing markets is very, very large, right? And
Anand Subramanian:image recognition has not been playing a major role probably
Anand Subramanian:for the past few years, but in last few years, we have seen
Anand Subramanian:like the technology evolving, especially recognizing products
Anand Subramanian:in very low light conditions and very products that that are not
Anand Subramanian:arranged well, products that are inverted, right? All sorts of
Anand Subramanian:complexities that come in, when when you go into smaller stores,
Anand Subramanian:traditioned rates stores, developing market, right, and it
Anand Subramanian:opens up a big, very big opportunity, because the number
Anand Subramanian:of stores that are huge. The value for the brands are also
Anand Subramanian:huge, because these are the areas where there are growth
Anand Subramanian:opportunities for brands as well, right. So I think image
Anand Subramanian:recognition is improving there. Definitely huge opportunity to
Anand Subramanian:digitize not just the large supermarkets and hypermarkets
Anand Subramanian:but across the globe, any type of stores, any type of
Anand Subramanian:conditions in which the store environments are present.
Angam Parashar:I think from a macro perspective, retail is
Angam Parashar:going to see a lot smarter demand forecasting and
Angam Parashar:replenishment systems, there's a lot of really smart companies
Angam Parashar:out there that use AI to really calibrate and get better at what
Angam Parashar:products should they be ordering, and you know, putting
Angam Parashar:in the back room and then putting on shelf and we can
Angam Parashar:reduce overall weight, shrink, etc. So I think with that,
Angam Parashar:there's going to be a very interesting dynamic from, you
Angam Parashar:know, there's shelf holes everywhere during the pandemic
Angam Parashar:to we might get better and better at closing those shelf
Angam Parashar:holes, because they're getting smarter with their ordering. I
Angam Parashar:think everyone that said there's a rapid technology advance in
Angam Parashar:image recognition at a lower cost is spot on that's happening
Angam Parashar:in front of our eyes right now. We'll continue, I would say to
Angam Parashar:advance in autonomous data collection methods. So less
Angam Parashar:people and more either robot or fixed hardware. And then I think
Angam Parashar:the last thing that's probably going to happen here is around
Angam Parashar:predictive analytics and we're going to see them really
Angam Parashar:increase in terms of getting really smart and calculated
Angam Parashar:about using a lot of different variables to do the next best
Angam Parashar:thing in store with the data that you already have. And I
Angam Parashar:think that that opportunity is just huge to kind of take one
Angam Parashar:big leap from where we are today.
Renish Pynadath:Yeah, this is Renish here again, I agree with
Renish Pynadath:what all my fellow panelists had to say, including the evolution
Renish Pynadath:of technology, which is gaining pace, lot more data available
Renish Pynadath:now, which makes it more actionable. One big, interesting
Renish Pynadath:thing that that we see happening is a lot of suppliers are, are
Renish Pynadath:able to move to what what we call as a store precision
Renish Pynadath:planogramming, or being able to kind of plan for every single
Renish Pynadath:shelf, in every single store, in every market that you operate
Renish Pynadath:in. Today, a lot of suppliers use common planograms across
Renish Pynadath:formats or in a large region, but we see that some of our
Renish Pynadath:existing customers are at a position with the amount of data
Renish Pynadath:that we are able to get them to be able to drive towards a
Renish Pynadath:planogram for every single store that they operate so like highly
Renish Pynadath:customized planograms based on the demographics based on the
Renish Pynadath:shelf based on their supply chain. So we are able to bring
Renish Pynadath:together a lot of data to be able to offer the ability to
Renish Pynadath:kind of plan for every single shelf at every single store. So
Renish Pynadath:and obviously other advancements like what Mike was, Matt was
Renish Pynadath:talking about in terms of predicting in terms of more
Renish Pynadath:automation in data capture, more automation in even in the model
Renish Pynadath:training, right. So today, AI model training is probably the
Renish Pynadath:is a heavy lift right now, how do we bring in more automation
Renish Pynadath:there, make that faster, make that more responsive? That's
Renish Pynadath:where we see this going.
Mike Graen:Awesome. Awesome. Did I get everybody's last kind
Mike Graen:of vision, I think I've covered everybody, Angam did you respond
Mike Graen:on to the five year vision.
Angam Parashar:I did actually.
Mike Graen:Okay, that's what I thought. You were the only one I
Mike Graen:wasn't sure about so. Well, we're gonna go ahead and wrap it
Mike Graen:up. We don't have any additional questions from any of the
Mike Graen:participants. I just want to thank each one of you for taking
Mike Graen:time, in some cases, early Saturday morning. Anand, thank
Mike Graen:you very much, go to sleep and you can wake up on Tuesday, if
Mike Graen:you want to. Appreciate you guys doing this. This is the future.
Mike Graen:I really believe this and I do believe automated data captures
Mike Graen:cameras, fixed cameras, robotics, but that that how do
Mike Graen:you take that information and turn it into insights so Matt,
Mike Graen:to your point, somebody can actually do something about it,
Mike Graen:I think is the future and I think I think the more we
Mike Graen:leverage this technology to make sure products are on the shelf,
Mike Graen:I think the better off we're gonna be. I thank each one of
Mike Graen:you for your time. Continued best of luck of your in your
Mike Graen:business and keep on developing, I think I think the future is
Mike Graen:very, very bright in front of all of you. So thank you very
Mike Graen:much. Have a great weekend and we'll talk to you later, bye
Mike Graen:bye.
Anand Subramanian:Thank you.
Angam Parashar:Thanks guys.
Mike Graen:Well hope you enjoyed that podcast on product
Mike Graen:recognition software. Please feel free to reach out to any of
Mike Graen:those suppliers if you have an interest or follow up
Mike Graen:standpoint. Next time we are back talking about RFID and the
Mike Graen:specific topic is the ID in RFID. What companies are taking
Mike Graen:advantage of that serialized item level whether it's in an
Mike Graen:RFID tag or a 2D barcode. The ability to uniquely identify
Mike Graen:every single selling unit is a key unlock for the future of
Mike Graen:technology in retail. Please join us then.