Aaron Conant  0:18  

Happy Wednesday, everybody, my name is Aaron Conant, the co founder and chief digital strategist here at BWG Connect. We're a networking and knowledge sharing group with 1000s of brands who do exactly that we network and now with share together to stay on top of the newest trends, strategies, pain points, whatever it might be that shaping digital, I spend the majority of my time just advising brands that start up the fortune 100. And sup every vertical would love to talk with pretty much everybody on the line today. If you're up for a great chat, that's where we get the ideas for these. Also able to help out a lot with service provider selection, if you're looking to going for any new service providers across the board, the network on an ongoing basis helps vet out the top people that are out there. So you know, I do a Google search, just put on my calendar more than happy to spend time with you. We're going to do close to 100 in-person events this year across the US if you're in a tier one city, please let us know we'd love to meet you in person. And a couple other things, just housekeeping as we get started, we're starting to few minutes after the hour, we're going to wrap up before the end of the hour as well, we're gonna give you plenty of time to get on to your next meeting without being late. The other thing is we want it to be as informational and educational as possible. So if you have questions along the way, please don't hesitate to drop them into the chat, drop them into the q&a or feel free to email me Aaron Aaron@bwgconnect.com. And we will we'll be able to answer them that way as well. I'm gonna open up the chat right now, attendees chat host and panelists. So that'll be good to go. And with that, I think it's a good time to kick off the call. So you know, kind of the topic here is around reviews and turning them into action. What we've seen is basically a leveling up of how companies have to use all the assets that they have. And it's different than even two years ago is different than even last year, customer acquisition costs have gone up. Conversion rates are more important, high conversion rates are more important than ever. And, you know, review is just false in that space. And what are you doing with those as a whole, we've got some great friends, partners, supporters of the network that come highly recommended from within the network over at Yogi and they agreed to jump on and kind of walk through how we should be thinking about this space as a whole. So Gautam. I'll kick it over to you. If you want to jump in and brief intro on yourself and Yogi, that'd be awesome. And you can kick it over to Spencer, and we kind of jumped in the conversation. Sounds good? 

Gautam Kanumuru  2:46  

Yeah, sounds great. Thanks for kicking it off, Aaron. So my name is Gautam. I'm co founder and CEO of Yogi so have been with the company since my friend in college, and I thought of it in a tiny New York City apartment like I'm in right now. So yeah, prior to founding Yogi, I used to work at Microsoft and was VP of Engineering at at a startup called collect on AI and I got acquired, I like to say if anybody here has any issues with any of their Microsoft products, I'm probably one degree removed from the person that kind of owns the product or features more than happy to send critical feedback their their way. And yeah, also kind of lucky enough to be a member of kind of Forbes 30, under 30 in the enterprise software and AI space. So yeah, that's a little bit about me quick, quick background about Yogi before passing it on to Spencer. So we are a reviews and ratings analytics platform. But really the way that we we look or the problem that we really see ourselves solving is almost a customer sentiment problem. Just really giving brands visibility into what consumers what shoppers really like and dislike about products, about brands about certain markets, giving visibility into that. So that's really what we're focused on. We live and breathe reviews and ratings as a data source every second of the day. So with that, take it away, Spencer. All right.

Spencer Kelty  4:10  

So I'm Spencer Kelty. I'm Head of Marketing here at Yogi. I've been working in the AI space for a while now I was head of marketing at constructor. And before that I was an agency strategist worked a lot of different companies to really help them along the digital transformation. I'm really focused on how cutting edge technology especially AI can not only help companies, but improve the customer experience overall, I think that as the case of the space gets more competitive as consumers demand more, they're looking for ways that technology can help them and give them a better experience with the brands that they like. And I think that brands are really catching up to that idea right now and trying to figure out how specifically AI can take them to where they want to go with those customer experiences. So we're really excited to dive in today, we got a lot of really great content for you. So let's let's get into it.

Aaron Conant  5:06  

And just really quick is the reminder, anybody has any questions along the way, like drop into the chat the q&a or email them to me, and we'll get them answered real time. Sounds great.

Spencer Kelty  5:20  

So the problem that we're kind of talking through and trying to address today is that consumer brands are always looking for more information. They're trying to figure out ways to find new data sources, to answer the questions that they have to figure out what shoppers want to learn more about the competitive landscape, ultimately letting them make better decisions and create better experiences for their customers, you know, across the across the board in all different industry verticals, this is what we're hearing that companies are really trying to do right now. Especially since the pandemic, shoppers have become a lot more easily distracted, a lot more likely to jump ship to another company, to change their brand loyalty changed their shopping experiences. And that makes it more important than ever to answer those questions about what's going to keep them with you what's going to bring new customers in what's going to make your product sticky. So that's really the problem that we're seeking to figure out some answers to today. And the goal here is to answer some questions. These questions take a lot of forms. What attributes do my customers like? What about my competitors? Customers? What do they like? How are consumers reacting to a change a new ingredient or a new packaging type? What marketing claims are resonating? And is there a mismatch in the expectations that I'm setting with my marketing and my PDP? These, a lot more can be answered by data sources like reviews. And that's what we're going to get deeper into as we as we kick things off here. So I'm going to kick it back over to Gautam here. Yeah, for sure.

Gautam Kanumuru  7:06  

And so when you kind of look at a lot of the questions that that Spencer was, was kind of alluding to in the slide before, things like hey, what attributes do my customers care about? What about my competitors customers? What marketing claims are resonating with with with our shoppers? This is one where, at the end of the day, what you really need to be able to accurately answer those questions is a great visibility into what actual shoppers are saying in pretty much as unbiased of a way as possible. And so that's what makes reviews and ratings as a dataset. Extremely interesting. In this case, when when we think of reviews and ratings, there's a lot of a lot of people, a lot of brands think about it in one paradigm. And they shouldn't be thinking about it in this paradigm, which is they're highly influential at the point of purchase, right? We want as high of a star rating as possible, make sure that we're outperforming certain competitors, because when somebody's sitting there and looking at two Amazon listings, they are more likely to go with the one with the higher star rating. And that makes that makes total sense. But on the other side, the when you think of what reviews and ratings are, they really are this unfiltered focus group of 10s of 1000s of actual purchasers who are pretty much it's almost like you're just handing them a mic and being like, hey, like, how was it, and they're just gonna speak their mind on various different things, they might talk about the taste of the product, the texture of the product, how it looked, how the box looked all over this kind of stuff. And so when you can really look at that information from that paradigm, there's a lot of useful information that you're going to pull out, they're going to answer those those very important questions. At the same time, pay, it's free and readily accessible, on demand and available as needed. This is constantly ongoing, it's not just a point in time kind of analysis. And we kind of alluded to it before, but it also gives you the chance to benchmark against competitors. Because everything that the same paradigm that exists for your products for your brands about what people are talking about also exists for competitors, and gives you the ability to be very in depth about that, that type of breakdown. And if we jumped to the next slide,

Aaron Conant  9:22  

I didn't want to I didn't want to launch this poll that we had, you know, this week kind of put this together and I'm gonna launch it right now if anybody can jump in and just see it. So essentially, is your brand currently doing review analysis? And yes, manually Yes. With analysis platform. No, we're not doing any. We'll just give this like 30 seconds or so as people can jump in. And they will end poll and kind of publish it to everybody. But I think it is like everybody wants to be doing it. But you know, we'll give this another 30 seconds here if you can jump in and fill it out, just I spend all day every day talking to brands. And this is Top of Mind. But the, it's just takes time and going through step by step, and everybody wants to do it. Everybody wants to feedback. Everybody wants to incorporate it into product design, feedback, detail pages, you know, answering questions, FAQs, whatever it may be, but you know, just give it 10 more seconds here.

Gautam Kanumuru  10:25  

Yeah, for sure. And we get and we'll, we'll, we'll, we'll dive into it a lot more. But at the end of the day, like, I think it's, it really is about one getting the process down correctly for being able to do it. And then you'll find ways internally to kind of scale it so that at the end of the day, the more that this is plug and play, the better it comes out.

Aaron Conant  10:45  

Awesome. So on my side, when I hit share, I'm not sure if everybody can see that. But did you guys see it on your side? We can sit on hours. Okay. I mean, it's it's pretty interesting. So 24% Yes, but they're doing it manually. The biggest one? No, we're not doing review analysis currently is I don't think any of those people don't want to. I think the reality is it takes too much time. And we don't have enough headcount in the reality is everybody's getting squeezed even more. I have teams being reduced in size. Yeah, super interesting here. Thanks for letting us drop this poll in here. I'm gonna stop sharing. Yeah, yeah, of

Gautam Kanumuru  11:30  

course. No, no, I think it's super interesting. I think there's also, one thing that we see is kind of all across the spectrum in in the way that people look at something like analyzing reviews, which is, what is sort of possible from what you can answer. Using something like this, right? On one end, it's, Hey, we're getting deeper visibility into a highly influential data source at the point of purchase. And so like, what can we learn from from that, but I don't think I've talked to a brand in the past couple of years, who doesn't have better just understanding the shopper or getting closer to the customer as kind of like a top three or top four consideration from from the top down. And when you really start to dive into and understand what people are talking about, and reviews and ratings to kind of Aaron's point, you start to be able to tease out how Hey, this is something that our marketing team should have visibility into, and that our product team should have visibility into. And so just really having kind of that, that deeper appreciation for what you're going to be able to tease out from this data source and how it's not just going to be influential on like the first bullet point making a change to kind of PDPs. But also, that that's maybe like a two month or three month change, versus something that could happen 12 to 24 months from now with, let's say product changes. And so I think that's, that's kind of an important piece, as well as really thinking about it from the strategic standpoint. And now, what does it take to kind of kind of get to that? So on sort of the next slide, what what we'll see is is sort of like the framework that very, very kind of simplified, top down mentality for how can we apply this in practice. At the end of the day, I think the, the, the reality, and especially when we look at sort of current economic conditions, downsizing routines, or limited resources is being able to make low hanging fruit changes, where the ROI kind of hits sooner, is much, much more important in today's day and age. And so at the end of the day, when you just look at the overall problem space, at the end of the day, what the the way that we look at this framework in a very simple way is you just purely start at the top, like hey, what is the data source or data sources that I care to analyze? And it's not just, oh, it's reviews and ratings? It's okay. What within reviews and ratings? Are we looking at just Amazon? Because that's a huge focus for us, are we looking at across different retailers? Are we just gotta focus on ourselves? Or do we just want to take in one key competitor, or, Hey, this is a hugely important market for us. So we're gonna go across the market and analyze everything that we can. So just being able to, to kind of get the scope correctly, and then choosing the right tools to analyze them, obviously, like, where we work at Yogi, we're obviously biased. We think we're the best tool out there. But for some of you, 

Aaron Conant  14:37  

that's good, right? Because, yeah, like we don't consider ourselves the best too. 

Gautam Kanumuru  14:41  

Yeah, then we have some more fundamental problems.

Aaron Conant  14:43  

I know we have enough people in the network recommended you so I don't mind that it's cool. This goes along with a question that comes in. I think it's in response to this poll for those of us who are not doing but those of us who are not currently doing reviews management, what's the best first step getting it started. So I think, again, everybody wants to do it. Right. But the reality is, what do I do first in, they don't want to go jump to step three, if there's three steps before that. And I guess that kind of falls into what you're saying here. Right is first in credit fraud looks like? We're all do I have reviews, ratings and reviews? And then which ones are important? Which ones are not? Which ones do I want to bring in? And then which ones do I want to analyze? And then are you getting actually into pulling out the analysis of the internal reviews? Yep,

Gautam Kanumuru  15:33  

yep, exactly. Yeah, at the end of the day, if this was like a day one journey, I think, where we've seen the most success, even kind of talking to customers, before they started using Yogi is really just start small, like pick your top three products that are on your mind right now, maybe the ones that are generating the most revenue for you, or maybe they're the ones that just came out three months ago, start there, pick three equivalent competitors. So keep it manageable, we're talking about kind of six products here. And just understand that the phase to go through is Hey, one, it's gathering the reviews. And then two, it's it's being able to parse through them to understand what are people talking about. And again, keep it simple. Maybe look at the top 10, top 20 most recent reviews across those just get a sense for what they are. And then the third phase is just being able to analyze them. So just understand, okay, this is what people are talking about, and go through the phase of sharing out the findings. So really, the the thing that always drives things home, this is something that we always recommend everybody is verbatim, like, literally just copy, paste, screenshot a review, sending it to someone is highly, highly influential. This isn't me saying, hey, when I read through reviews, I noticed that our taste is underperforming. It's, hey, know, Bill 123 on amazon.com had this critical feedback, guys, I think we should, we should look into this. And so that's really where it kind of step one starts. But the other phase of it when we kind of look underneath of what this causes is, drive a business change from it, that that's the key, I think a lot of insights tend to stop at, wow, this is super insightful. Wow, this is super interesting. Like an email thread that was like, Oh, this is fascinating to me. And then it just just dies off. So really tying it to, to a business change. And the one that we actually recommend starting with, especially in this environment, is a PDP or digital campain optimization. The reason is, is because the turnaround time on this is a lot faster than a marketing claims change a product update, new product innovation, those are we're talking 1224 36 months down the road, but a PDP change in theory, you can make it in a few weeks. And so if we say if we see like, hey, it seems like people are talking about tastes and 80% of our reviews. But in our bullet points, it's like the fourth bullet point. And in our images, it's only talked about on the fifth image than literally just reordering that can have downstream changes. We're talking about even just a point 1.2 increase in conversion rates that lead to obvious ROI downstream. And so I think that's the way to really look at it is to start small. If this is day one, like tools like Yogi, you probably it's not the time for them, right? It's all about just kind of building the muscle for it. And then showing that you can drive change from it. Before you you really kind of commit to it on a grander scale.

Aaron Conant  18:31  

Yeah, one thing, yeah, jump in Spencer, then I have another question I get to but I want to hear your thoughts on this as well.

Spencer Kelty  18:36  

Yeah, no problem. One thing I just wanted to add to that is, you know, let like Gautam said you got to start somewhere. And I think that even before you start pulling in reviews, there's kind of a couple different ways you can you can think about it conceptually, you can say we just want to start looking at reviews to understand customers better. Or you can come at it with a specific question in mind, and let that inform the way that you go about it. If there is a business problem that you're experiencing, if at one certain product has a very low conversion rate compared to your others, if one products underperforming, has much lower reviews, that can help you figure out where to start. If there's something that's a specific focus, that's going to help you narrow it down, you know, if you have one product in particular, that's, that's a problem product that isn't performing the way you want it to. Well, there you go start with that, look at your reviews there pull a couple of competitors that are performing maybe closer to how you wish that yours was performing. Or vice versa. You have a product that's performing really well and you want to learn more about why it's performing well compared to the competitors, pull reviews from that and start there. And you know as as Gautam the set, it's really about kind of looking at that product lifecycle. Look at your roadmap and figure out how you can make actionable impactful changes over time. KDP is fast that has a you know about a point one star rating impact associated with it. We seen that from a ton of different companies that have taken these insights and within a few weeks action changes that resulted in that point one star rating. Now, you know, I noticed some people that might not sound like a lot, but I'm sure to everyone here who's in products every day and obsessing over every little detail. Point one is a lot. And that can trigger those bigger changes, too. Because when you start learning and start looking at those reviews in depth, well, you're going to have ideas now about how to change your marketing claims. If you reorder the way that you're talking about your product and your PDP, and you see your conversion rates go up. Well, that just told you something about your audience and about your market. You've now seen the feedback, you've tested it, you've gotten the results. Now you can apply that in marketing, you can apply that even in your product updates, iterations. And that's where you see those, those bigger jumps over time. Obviously, those take longer, but you can see we see an average about a point to increase as those changes go. So that averages out to a point three star rating increase on a product from just using reviews into its lifecycle.

Aaron Conant  21:07  

That's awesome. Yeah, no, you answer the question was around are you using this on competitors datasets as well? But you are in how are you seen brands? So they were like next level? So I can see you're doing on your own products, and you're updating your PDP. But are you also do you have people in brands, I can see them crunching the competition's ratings and reviews. And if they see the same thing that they've had tastes as an issue, then you bring into yours, you might not have anything on yours that says taste. But if you've seen some complaints on taste, and yours, you put, you know, new, better tasting formula or something I don't know. But do you see brands using it that way as well, this is next level is going again, outside of what the norm has been used in the past? Absolutely.

Spencer Kelty  21:54  

I'll Gautam going a minute, but I want to I want to kick it kick off the answer to that by saying that we've seen some brands even take it a step further than that we've seen brands before a product release or with a product release that doesn't really have reviews, going and looking at their competitors, looking at their competitors reviews, and using that to build their own product Strategy. So it doesn't even need to be a product that you have a ton of reviews for. It can be something that you're looking at the market and trying to get a launch point, based on what everybody else is already doing in the feedback they've gotten. Let them make mistakes, let them inform the market. And you can come in and be the hero after that. And that's something that we've seen quite a few of our companies do. And to great effect, you know, if you can go and look at all of that feedback, you can jumpstart your product lifecycle. By miles, you can go ahead and do things that would have taken you two, three years of learning to figure out. And I don't think that's anything extraordinarily different than what a lot of companies do already. But like kind of going back to what Gautam said a while ago, it's kind of going from that one and done mindset of doing a focus group doing testing, to an overarching philosophy around taking feedback and applying it consistently. So it's kind of that paradigm shift that we were talking about. And I think that the more standard application is somewhere in the middle, it's you have your own reviews, but there's a big market out there, people are talking about other products, you can learn just as much from that. If, if a competitor goes and changes their packaging, and it's a packaging option that you are considering, and you see that their reviews, take a dive that reviews that mentioned that type of packaging have lower sentiment? Well, that's a pretty good indication that you shouldn't do that, that you should probably take that off of your product roadmap. So the limit is really just to the questions that you're trying to ask and trying to answer. And the actual data that's out there, like are your competitors getting reviews that are helpful? And if they are, you can absolutely use that. Sorry, kick it back over to you Gautam didn't mean to take it from you.

Gautam Kanumuru  24:10  

No, no, it's all good. I mean, at the end of the day, I think the way that I would like simplify it is it's a question of context at at at the end of the day, to the point where I would argue, hey, if we only have the bandwidth to analyze six products, you're better off doing three of yours and three competitors versus six of your own. Because I can't tell you the amount of clients that tell us, hey, we saw like we whenever we dive into our reviews, we saw complaints about x some complaint about our packaging, but when we were able to benchmark it with the rest of the market, it actually turns out we're still doing better than everybody else. So a one off review of somebody complaining that Oh, their packaging stinks can cause a huge red flag and a bunch of innovation and a bunch of Dawn 1000s if not millions of dollars invested to make the change, when in context, you're actually still doing better than the market. So you're better off using that money for something else. So that's kind of like the key point with with looking at competitors, I would say it's a, it's a must have from from this perspective, 

Aaron Conant  25:18  

yeah, when we started this, I didn't think of this product as what you have is a must have. But it's, you're saying, If I do any product design, if I do any product launch, if I have any competitors out there, right? I mean, think about product design standpoint, right, you're not only crunching yours, but all the competition and everything that's out there that anybody could look at, you're crunching the data is kicking out, hey, this one is too wide. This one is not the right color. You know, the wheels are too big, the packaging is too heavy. And it's not just product design is packaging design as well. Like you say that it is going to speed up product development, and is going to increase the ramp up time for new product launch, which is the hardest that anybody has to experience right now. Right, as you get a new product out there, you're on page 5000. Like you don't have any ratings, you don't have any reviews. Ah.

Spencer Kelty  26:18  

And I think one thing to add to this

Aaron Conant  26:20  

one, I did not see this. Yeah,

Spencer Kelty  26:23  

this from a marketing perspective to you, if you're looking at your competitors, and doing competitive analysis, like reviews are such an amazing tool. You know, as a marketer, I'm sure we've got marketers in the audience who can who can relate, it's so easy to spend so much time obsessing over the competition, looking at what they're doing, looking at the claims, they're making, the way that they're talking about their products. But with with reviews, if you have a good review volume and analysis Strategy, you can go and look at how a competitive product is positioned, and then look at the reviews and treat it just like you would your own. Basically, pretend it's your own product, and decide what changes you would make sure the marketing Strategy based on the marketing Strategy you see in place for that product, and what the reviews are saying, if if a brand is marketing their their product, as let's say, a health food product or a natural product, and there's complaints that it has too much sugar, well, that's a mismatch, you know, you can see that you can take those learnings back and understand that the audience is sensitive to those things. It's really just just a matter of thinking about it as a global marketplace. And you're not just caring about your product, you're caring about what everybody's doing and how you fit in.

Aaron Conant  27:44  

You're awesome. So I have a question comes in from Kay here. How does the platform actually work? Which I guess we should probably get to, but usually, we'll we'll stay away from like the sales pitch, but I don't mind you, you know, walking through how it works? And then can you please talk more about how you're gathering the competitor product reviews? And how you're discovering the context of the reviews? Right, for instance, you know, going from knowing a product is underperforming, say two stars and identifying how the consumers are telling us the product can be improved. So yeah. What's the internal guts of of of how this works? And and how do people implement it? And then we can get to the rest of the deck? Yeah,

Gautam Kanumuru  28:27  

yeah, for sure. At the end of the day, I'll try to answer kind of like both both questions at the same time. And I think we actually might might have a slide on this. So I'll give a kind of like high level overview. And we can we can dive in a little bit later. But Yogi works in kind of three phases. The first phase is what we call aggregate. And so this is building the a central repository of the reviews and ratings data. And so pretty much at the end of the day, for anybody that's tried to like manually sort of analyze reviews before, it is actually a pretty massive headache to go from here are the products I want to analyze to here's all the sources that they that they exist on. And so we've built out a way to essentially be able to create those product mappings. But but pretty much just think of it as we build that database of reviews and ratings. And it is it consists of some combination of kind of like crawling different retailers finding the relevant URLs and then pulling the reviews from there. And so that's sort of the first phase. Then the second phase is what we call organized. This is where we use our AI and natural language processing to break down the actual text. And so we're not talking like a key words or word cloud or something like that. What we'll actually do is recognize the different topics of conversation that people are talking about, and label them accordingly along with a sentiment so I can't tell you the amount of five star reviews that come out that's like oh, this Is amazing best thing in the world. Like the like, I can't live without this, I just wish there was a blueberry flavor or something. So even in a positive five star review, there will be kind of like some negative mentions that are highly, highly valuable. And so that's what that layer takes care of. And then the third is analyzed where we have kind of a browser accessible dashboard to give visibility into that. But we can dive further into that, but to the point of competitive analysis as well, when we talk about that aggregate layer, we also take care of competitors there too. And then yeah, Kate, to your to your other question about Bazaarvoice. Yeah, they become one of the sources that we're able to pull from.

Aaron Conant  30:42  

Awesome. Yeah, we'll use them Bazaarvoice today, so they had pulled in from Bizarre Voice and crunched.

Gautam Kanumuru  30:47  

Yep, yep. Yeah, that becomes like another source for us, essentially, is Bizarre Voice. But yeah, I know, Spencer, you had some kind of case studies that we can that we can walk through that I think will give some good examples of this in action.

Aaron Conant  30:59  

And thanks, everybody, for the questions, keep sending them in chat q&a In emailing them. And we'll keep getting an answer.

Spencer Kelty  31:05  

Awesome. You know, over these next four or five slides, we'll get deeper into some of the some of the nuances in those questions, too. So just stay tuned. Yeah, the first one I want to talk about here is Tylenol. Johnson and Johnson and Tylenol are is one of our biggest customers, we've worked with them for quite a while, they came to us with a product that wasn't performing quite the way they wanted. It was a newer product launch, Tylenol dissolve packs. They were trying to get deeper visibility into into why what was different about this product from their top performers. And what they were able to discover very, very quickly in the yolky platform was that a huge portion of their negative reviews and their reviews that had negative sentiment was due to a direct misunderstanding of how to use the products. People see a little packet of powder, and they think they put it in their drink and stir it up and drink it. But this product was meant to be poured directly in the tongue. And they had the instructions in the box. They had all of that. But a significant portion of users were just thinking they understood how to use it, ripping it open, pouring their bottle of water and wondering why it wasn't dissolving. Why was it not working?

Aaron Conant  32:28  

It literally says dissolve packs on it, though.

Spencer Kelty  32:33  

Yeah, they were they were, you know, making a somewhat understandable conclusion there because they'd been trained by other products like, you know, liquid IV and things like that, that you have a little packet of powder, you put it in your water, I believe it was about 90% of one star reviews, at one point were traced back to this. So what they discovered with this is that by changing some elements in their PDP, and then later in their, their messaging, I think even their packaging got a few changes based on this information, they were able to aggressively counteract that. And there's there's another thing too, like that's kind of a negative issue with the messaging that they were able to correct there was also a positive they were able to jump on and lean more into, and that is that they were finding a high number of positive sentiment reviews that were mentioning fast acting, which wasn't part of their original claims, when they launched the product, they didn't consider it to be part of their fast acting line. They weren't marketing it, or messaging it as a fast acting product. But a significant enough portion of their audience was reviewing it and saying this is my favorite channel or product because it works fast. They were able to go in, add that messaging, add that marketing in and you know, lean more into what was working for their customers. Now for the results, because they're massive, they had an overall plus one organic review star rating increase. So the product went from well below average to that for them. As far as ratings go to Above Average. That plus one organic star increase was a massive change. And it spurred a 9% increase in online conversion rates. So this isn't just a customer experience thing. This isn't just a thing where you make changes and customers get a better sense of the product. It has a real impact on your star ratings on your conversion rates and the entire shopping experience. So this is one of the coolest examples because they were able to quickly find both a negative they are able to to fix and improve on and a positive they're able to capitalize on

Gautam Kanumuru  34:52  

Yep, it's also a good example of a low hanging fruit change sac. So that's, that's super important I think the next one was jumping into is, is with Nestle. And so this is an interesting one where we're not looking at kind of the three month time horizon, maybe more like a six to nine month time horizon. But one of the things that they found using Yogi was actually kind of an opportunity to fill a gap in the market. So what it turns out is that they were starting to using Yogi what they're starting to see in reviews is more and more mentions of people using coffee creamer, because they own the coffee mate brand, not in coffee, people are using it in ice cream and hot chocolate and teas. Even some people using it for baking, for example. And so this, they saw this as an opportunity to kind of grab a new segment of customers that they potentially never targeted beforehand. And so what they're able to do is start to introduce this from kind of like a marketing perspective. So if you see this image, you'll see kind of its fall themed with with more of a like a like hot chocolate, versus a coffee or something like that. And they actually started to kind of take control of this this messaging, if you will. So this inherently drove more increased mentions of coffee creamer, coffee creamer being used and not coffee. And it allowed coffee made to kind of become like the leaders of it. So this is a this is more of kind of finding a gap in the market and taking advantage of it. But I think the interesting thing to note about like Spencer's what Spencer was talking about with Tylenol and this Nestle case is it's the same data source, it's the same type of analysis. But when you look at it from two different dimensions, you're able to tease out completely different answers. Yeah,

Spencer Kelty  36:54  

I think that, you know, one thing I'll add to this one that I think is really cool. From a marketing perspective, with a brand like Nestle and coffee mate, it's ubiquitous, you're going to be hard pressed to find someone who doesn't know about coffee names. But for them to be able to go and find new markets, or for a product that's so ubiquitous, is huge. So 60% increase in mentions of these other beverages is a big growth area. For them, it became a big focus. Because, you know, you're only going to be able to do so much expanding stronger to the coffee market, when you're the leader. The big opportunities are in these sorts of kind of adjacent market expansion. So this has been going

Aaron Conant  37:41  

on forever. You're I mean, they just found this out in the past, I don't know a year or so isn't that unbelievable? Like, yes, like all of a sudden, people started using it in tea or hot chocolate or to flavor other beverages, they've been doing it forever, they just never been marketed to?

Spencer Kelty  37:59  

Well, I think that one part of that, too. That's, that's interesting is, and I think probably a lot of people can relate to this, it's maybe they did, maybe they did have a sense of it. But it's hard to, to navigate making a product decision and a big business decision to focus on that if you don't have a value associated with it, you don't have data associated with it. So that's kind of one of the things that we'll talk about a little bit later. But when you're talking about kind of a manual analysis process versus something like like Yogi, you know, when you're doing it manually, you probably are going to be more gut feeling, you're going to say we surveyed 100 reviews, and we found these, you know these things in them, versus we surveyed all of their reviews for all our products, all our competitors, products, everything in the industry, and we're seeing this data trend, that becomes a lot easier to make a business decision like this to focus more on another market. You know, when they may have known like I, I'm going to pretty much guarantee that the marketing team and maybe the customer success team people that when they're talking to customers day to day, they probably knew that a subset, we're using it for other things, but there was no data behind it to make a business decision to move forward.

Aaron Conant  39:12  

Yeah, I don't. I'm just I'm just sitting back thinking, how can you not have some? How can you not be looking at these every day? And I've not started to call thinking this at all, because you're talking about your mining ratings reviews for not only new product launch, but product design, also competition, also marketing and also potentially identifying new markets, or marketing demographics. It's all from the ratings and reviews. It's it's unlimited, nearly free focus groups on an ongoing, updated basis. Not all of your products. Yeah, exactly. I mean, no It

Spencer Kelty  40:00  

goes back to the paradigm shift that, you know, both both Gautam and I touched on earlier, it's, it's going from a kind of one and done customer feedback and insight cycle where you'd go and do a focus group, you know, every six months or every year and get that information, which, by the way, from the information we're getting from, from the market, focus group budgets are going away like that r&d budget is, is shrinking all the time. So it might have been once every quarter now, it might be once every six months, once every year that a team has the budget to go and do that. Something like this. It's not one and done. Anytime you have a question, you can go in and get the answer that day, you don't have to spend that time to organize a focus group, wait for that data to come back, wait for your analyst to look at it, you can go in and find it immediately. So and that brings us nicely to this. I'll start and then I'm sure Gautam. I want to jump in here. But you know, when when this goes back to some of the questions that we had earlier, when we're talking about taking action and getting started on building a review Strategy, and a review analysis Strategy, there's kind of two ways to go about it. There's the you know, let's let's dip our toes into it. And let's jump all the way in the manual analysis is kind of the the way to start, if you want to see some value, you want to figure out what you can get out of it. But there's definite limitations. So when I'm talking about manual review, I'm kind of talking about that process that Goeth was talking about earlier, saying, let's pick a product, let's pick a couple competitive products, let's go and, you know, pull the last 20 reviews from those products, something like that, you're figuring out a limited sample size, you're figuring something that's manageable to do by hand, and you're pulling those in, likely into a spreadsheet. I know we always love having more of those in our lives, pulling those into a spreadsheet or something and then basically sitting there and using maybe using your Excel sheet to you know, chart a graph or something like that, or even just using you know, Ctrl F to go in and find keywords that matter to you. You know it, it certainly can answer some questions for you, it can provide some insights, but it's a little bit like going into a dark room with a flashlight, instead of just turning the light on, you're gonna be able to answer some specific questions, if you know what you're looking for, and you're willing to spend the time to find it. It can also provide some some easy answers like, you know, have our reviews improved in the last three months, like that's a relatively easy answer to to come up with from manual. You know, you can also say that, you know, let's go in to one product and pull all of our one star ratings, you know, and run a run through some some common keywords that we see and see which ones pop up the most like, those are really just a function of how much time you're willing to put in. But they don't give you the kind of the kind of depth that something like a Yogi solution would an AI powered review analysis where you're pulling in every single review of your product that exists in all of these dozens of sources across the web, you're pulling in the entire competitive landscape, you're pulling in the entire industry vertical, usually, to bring all that data inside, so that you can benchmark it against every competitor that is in the space, every competitor that enters later, can pull that data in, and basically flip that light switch on to get all the data in house immediately. Now one thing goes and went over briefly earlier that kind of the pulling in the the organizing of the reviews, and then getting into analysis. One thing that I think is really important, too, is to talk about how the metadata is still there. And the metadata survives through all this. So what I mean by that is everything still there on where the review came from, when it was written. Any demographic information that's included on the site is already there. Like is it a verified buyer? Is this a sponsored review, like all of that still exists? So those are all filters, all things that you can look at and get deeper into. So you can just say like, you know, let's plot something over time. Like it down here. In the example questions I've got how this time of year impact consumer opinions on our product sugar content, like that's incredibly specific. That requires filtering by time of year filtering by sentiment around sugar content. And then you can even take that and benchmark it against your competitors. No other platform on Earth can do that right now. You know, Yogi is the only one that has that level of in depth analysis. And those are the type of questions that companies really are trying to answer. You know, if you're trying to create a product that appeals to a kind of health conscious audience, and you'll know that people get more health conscious right at the start of the new year, they start thinking about what they want to change in that coming year. And you'll see changes in sentiment around things like ingredients. I've been talking for a while, I want to let Gautam  jumping on this slide as

Gautam Kanumuru  45:13  

well. Yeah, no, I mean, I think Spencer kind of kind of hit on a lot of it that the thing I would say is, like, for, for somebody who's who this might be the first time, like thinking about reviews from from this context, or like being able to, like see it for yourself, if you can extract as much value that's really where like, starting with the manual review analysis piece is a good way to go. Like, you're, you're gonna start to be able to tease out like, oh, look, and actually a pretty data driven way, I am able to get to a good answer of this. And that's when you truly start to understand the power of it. And that's when something like a Yogi tool, AI powered review analysis becomes sort of like justified if you will. And so it's definitely at the end of the day. The the way it when when you really start to look at the questions that we're able to answer, and we've been given examples kind of throughout the presentation, it's almost interesting to look at it from the perspective of what's truly being given visibility here is shopper sentiment. We just happen to be doing it from a highly powerful data source, which is reviews and ratings. But at the end of the day, really what it fundamentally is, is shopper sentiment visibility, which is like highly valuable across an organization.

Aaron Conant  46:54  

This is awesome. Awesome. Yep.

Gautam Kanumuru  46:58  

Yeah. So we can, we can definitely want to be conscientious of everyone's time. So we can just like jump jump to these very quickly. I know kind of Kate had a question of how things work. And so this is what we were talking about from the aggregate organizer. When is there

Aaron Conant  47:11  

like a trial? Like how do people try it out? That's the other question that comes in.

Gautam Kanumuru  47:15  

Yep. Yep. No, no, I think it's, it's a good question. So at the end of the day, like, the best thing to do, if you kind of want to see this live is really just reach out to us, one of the things we're always happy to do is just run some of your data through through yogi. And so from there, you'll be able to get like a truly deep understanding of hey, like, what can I truly answer from this? And so so that

Aaron Conant  47:40  

we can connect, everybody will shoot out an email afterwards? Because it's worth if you're going to crunch some data for free, why not do it? The other question comes in, what's the spectrum of review sources that Yogi pulls from?

Gautam Kanumuru  47:50  

Yeah, so I think at this point, my engineering team always boasts that the number of retailers we're able to pull from I think we're at like, 450, Intel as we pull from so the relevant ones, right. Okay, Exactly, yep. And we go international. So we cover every major country, every major language, at the end of the day, the best way to think of it is, as long as the reviews and ratings aren't hidden behind a login, we're able to pull it in on our end. Awesome. So yeah, so we can probably just kind of like end end on on this example. But I think this is a good one, just painting the overall picture, end to end. So if you kind of look at the top part, you'll actually see that this is the same exact review that actually shows up on four different sources. So one of the things that comes with reviews and ratings data is it's actually inherently messy. So this is a concept called syndicated reviews, which means review, the same review can be shared across different sources. So that's one where if you're just doing this manually, like reading things, you actually might end up double, triple quadruple counting the same data point and obviously, that can that can screw up your analysis downstream. So that's one thing that we take care of, on top of other things that come with reviews, like promotional reviews and, and stuff like that. Then the second piece is organizing it going through that text and actually break being able to break it down, not only on a sentiment basis, but also a topic of conversation. So you can see things like the way the instruction set to change it turns it into goop. Change it back, horrible instructions, great mix. So you can see here that this chewy fudge brownie mix product is not so much facing issues from the actual product itself as much as the instructions that they're providing people to follow in order to build the product. And now when you're able to do that across 10s of 1000s of data points, you're able to track sentiment over time, which that graph in the bottom left shows as well as get really in depth in like breakdowns 22% reduction in sentiment 40% 43% increase in mentions of instructions and how you rank in the overall market based on that. So That's probably the the example we can kind of, kind of close with with folks of this is just on one product, maybe even across one retailer, but hopefully paints the example of what's possible on a grander scale.

Spencer Kelty  50:13  

The one small thing I want to just emphasize there is, this is a visual illustration, in the center of what the AI does, like that's the AI part of Yogi is, the machine learning algorithm looks at this text data and turns it into quantitative information. So it's pulling out the key parts, the most important elements of this review, and assigning positive or negative or neutral sentiment to it. So horrible instructions, it pulls out user experience in the instructions category is negative, it pulls out the fact that this person has positive sentiment around being a repeat purchaser, it pulls all that information out. So you don't have to look at that individual review. To get this information, all of that information is put into the aggregate so that you can filter in for any of those individual attributes or features, and see how people are talking about it over time. And that goes down to that bottom graph where we look at overall sentiment for the product over time, it took a massive dip after that instruction, or after that recipe or instruction change. And yeah, we can see, you know, like a 22% reduction in the sentiment for the overall product. But you can also go in and look at how it changed around instructions, or texture or recipe or anything that is relevant to your product that reviews

Aaron Conant  51:37  

talk about. Well, this has been awesome. I'm gonna have to have you on the podcast for sure. This just an educational podcast where we're tackling these issues and people can digest it, I just put a link to everybody. You know, it's everything from what's going on with Amazon to how does composable commerce and architecture work. But also, we have to do a deep dive on this around ratings and reviews and just from a Strategy standpoint, because this has been fantastic. So encourage everybody, you know, we'll have these guys on the on the digital deep dive for sure. But thanks again, Gautam Spencer for your time today. Thanks for everybody for sending in the questions. Look for a follow up email from us. I would love to have a conversation with you know what your pain points are. So we can tackle those on a future webinar or maybe an in person event. But also more than happy to connect you with the team over here at Yogi it's the 100% worth having that connection and just having them run some data and see what they can do. And with that again, Gautam. Spencer, thanks so much for your time today. Thanks everybody for dialing in. Hope everybody has a fantastic Wednesday everybody take care, stay safe and look forward to having you in a future event. Alrighty, thanks again.