3488: How Akeneo Sees the Future of Product Experience in an AI First Retail World

[00:00:04] What happens when AI knows what you want before you even think about clicking search or hitting a search engine? That's the question that's shaping the future of online shopping. And from playlists to grocery lists, I think recommendation engines already know how we spend our time and money. But yeah, trust still remains the missing ingredient in this new world of AI-driven commerce.

[00:00:31] So the question I want to ask you all today is can technology that predicts our preferences also earn our confidence and our trust? Well, my guest today, I hope has the answers. He's the CEO of Akeneo, a company at the heart of this transformation. And I want to talk about what 84% of consumers are getting right about AI recommendations and why chatbots, well, they still have a lot of work to do.

[00:00:58] But most importantly, how product experience, not just the data, could define the next era of digital retail. So whether you are leading an e-commerce team or just curious about where your next online purchase will come from, or if you are a retailer, how you're going to reach your audience as they turn away from traditional search engines and replacing by chatting to AI assistants,

[00:01:25] I'm hoping that the whole conversation will make you think a little bit differently about how AI is already shaping what we buy and why we trust it. But I'd just like to pause for a second to thank our sponsor for helping me share these stories with you. Because keeping the Tech Talks network running and producing over 60 episodes every month takes real support behind the scenes. And I'm genuinely grateful that they've helped make this possible.

[00:01:51] If your business operates in a regulated industry, you already know how important compliance is and how painful it can be to maintain. Now, NordLayer, they're making it a little bit easier because they're providing audit trails, encryption and access controls that help you stay compliant without building everything from scratch. And you can monitor device posture, enforce policies and prove that your data is protected in transit.

[00:02:22] And for industries where privacy and trust are everything, this can be a huge relief. And because it's a cloud solution, you can roll it out in minutes. So if compliance has been keeping you up at night, let NordLayer take the weight off your shoulders. Visit nordlayer.com slash techtalksdaily and save 28% using the code techdaily-28.

[00:02:48] And on that note, I think it's time for me to officially introduce you to my guest today. So a massive warm welcome to the show. Can you tell everyone listening a little about who you are and what you do? Well, it's great to be here. Thank you for having me. My name is Roman. I am the chief executive officer at Akeneo. Akeneo is the product experience company. We help companies make sense, manage and distribute the information about their products

[00:03:17] just simply to maximize their sales, do where they're returns and get ready for the massive transformation coming in commerce that is due to AI. Well, thank you for sitting down with me today. One of the things that put you on my radar was when you set off my tech spidey sensors. When I read that your data showed that 84% of consumers who made a purchase based on an AI recommendation were happy with the experience.

[00:03:43] I've got to ask, looking at that survey, that report there, what do you think is driving this new level of satisfaction? And how can brands and business leaders listening maybe build on this momentum? Well, that's a big question and a big topic. But it's crazy to see how fast the numbers are evolving when it comes to AI-supported commerce. And at its heart, you know, why are people so happy? Why 84% of customers expect satisfaction? I think it's just because it works.

[00:04:13] You know, we've heard so much about AI and data for so long and struggled to actually get the benefits of it. Here, consumers do get the benefit. It makes their life simpler. It makes their life easier. It makes it just that much easier to find the things that you need and the things that you want and to drive you through that whole experience. So I think the big shift that is coming with data and AI, it is now making a change for the best in the life of consumers,

[00:04:41] both B2C and B2B, by the way. And of course, AI recommendations surround us everywhere we turn, whether it be Spotify knowing what album or song we might want to listen to, or Netflix knowing what we want to watch, or Amazon knowing what we want to buy. But just to keep grounded here, on the flip side of what we're talking about, I think it was only 45% of consumers said that they fully trust AI-powered recommendations or chatbots.

[00:05:10] So from your perspective, what are the practical steps that retailers could take to maybe earn that missing trust? There are a couple of steps that can be taken. But I think first and foremost, the fact that it's already 45% trust in AI chatbots, even though they've only been here for a limited amount of time, is in my view huge. Because ultimately, the only way to build trust is to actually use it and to see where it takes you. So I think 45%, if you had asked the question a couple of years back,

[00:05:38] I'd be very curious as to what you would have gotten. So the first factor to actually drive that is just a bit of time and a bit of use. And if you see how fast the use of conversational AI is ramping up, I wouldn't be surprised if that's one of the biggest factors in increasing it. Then more on what the brands and distributors can do, it's fairly simple. It's if the AI doesn't know better, it's going to say stupid things about your product, and it's going to erode trust.

[00:06:06] So how do you make sure that you fit this system with data that is accurate, data that is trustable, data that is comprehensive? And ultimately, you know, I've been working in AI for basically 25 years. Before it was cool, I want to say. I'm not sure it's cool now, but before it was cool. And the one thing I heard my customer tell me over and over and over again, it's, I love what you're talking about, Roman, but my data is just not good enough. And that's the actual secret of AI, is there is no good AI without good data,

[00:06:36] which means there's going to be no good AI-based commerce without good product data. And I think to be ready, understand that this is the biggest challenge that companies are facing to date. How do they get their product data in order so that they can actually power this AI-based application is one of the biggest topics of the month right now. And then how can they actually get that data in order? Well, we're here to help, of course, but I think it's a company-wide initiative

[00:07:04] to make sure that you will have what is needed to provide the right information so that your consumer can have trust in what the conversational AI agents are going to share. Now, one thing I often tell to people I talk to on the topic, it's everybody's scared about AI hallucination. But if you give it wrong information to begin with, what if it hallucinates? Maybe you had a chance it hallucinates the truth, but most likely not. So just make sure that your data is in order,

[00:07:32] and then you have a chance of building trust for the consumers. Yeah, I completely agree. And I think two out of the last three events that I've attended, I've used that exact same phrase, without data, there is no AI. And maybe it feels like there's a certain gap between consumers enjoying AI recommendations and their maybe hesitance to engage with chatbots due to some bad experience before AI came along.

[00:07:59] So why do you think chatbots are struggling to gain that same level of acceptance? And how can we change that as well? As I said, I think on the chatbot side, there's a big question of just the ramping up of the usage. I mean, the numbers are staggering, right? I think there's been a 50-fold increase in less than 12 months on the use of these systems. Of course, OpenAI released by it in ChatGPT.

[00:08:24] All of the big LLM players are looking at how they're going to include some elements of commerce into their activities. So I think ultimately, you know, the technology chatbots didn't really work until a couple of years back. Now they've started to work. Now they're being used for commerce. And the ramp up is just massive. And every company in the world, I'm sure, is going to be using it because it provides a completely different type of service

[00:08:50] from what our keyword constraints experience has been as of to now. When you want to buy something, you kind of need to think about your problem, then identify by yourself your solution, then turn that into keywords, then search, then browse, then go. Whereas in an environment like a conversational AI, you just type what your problem is and you might just find your solution.

[00:09:17] And transparency, of course, is another reoccurring thing that we're seeing in AI adoption this year. So how much do you think customers actually want to know about how their data is being used? And where should brands draw the line between openness and complexity and avoid drifting into creepy territory? Yeah. So that's always the difficult thing, right? It's how, first, the regulatory aspect. How do you make sure that you comply with local rules?

[00:09:46] You know, in Europe, there's GDPR. Many countries have laws emulating it. So that's, of course, a big element. Then there's indeed how do you do not get into creepy territory? Because, you know, you're happy to have... If you go into a shoe store and the person talks about the style you're wearing, how you're dressed, it feels good and people will be happy to have recommendations. If the shoe seller starts saying where you live and your phone numbers, then it's a bit scary.

[00:10:15] It's the same with AI and with LLM. You want to make sure it's the right level. I think that balance is starting to be struck. Companies are understanding what's important or not. But more importantly, I think the evolution of the way people buy is making the reliance on creepy personal data less important. Because ultimately, one of the challenges of people today, of sellers today, was to kind of infer the intent and the profile of their buyers so that they could offer the best product possible.

[00:10:45] Once you start with conversational AI, since the people actually start with their intent, it's easier to understand what people want. So you have less inference to do and a bit less reliance on personal information. So my hope is that these new technologies will actually help sellers, distributors and brands need a bit less personal information because they can rely a bit more on what their product are actually going to be used for. Which is better for everyone. A bit more privacy, but a lot more relevance.

[00:11:13] And one of the things that made Akenio stand out for me is how you focus on product experience rather than simply product data. So how does that distinction help brands design AI experiences that feel maybe more human and more trustworthy and everything that we're talking about here? Because, yes, we talk about product experience because that's what we're thriving for, right? We want to allow brands and distributors to offer the best experience possible to their consumers. And that's what we're having for.

[00:11:43] And to do that, you need to have product information and to have product data. It needs to be comprehensive. It needs to be as fleshed out as possible. It needs to be informed by many elements from your suppliers, the regulators, also feedbacks on what is being shared. And so we see a role not just to help generate, govern product data, but to make sure that these guests enrich and align with all of the rest of what comes in a product experience.

[00:12:11] And more importantly, that it flows from system to system in the most efficient way possible. I deeply believe that ultimately it all comes full circle of you have a product experience that is partly product data, partly assets, partly the expense itself that then get exposed to different channels or get to different channels to the end consumers. Then that allows you to get feedback on performance, on satisfaction of customers.

[00:12:40] And it can be funneled back into the product experience environment so that you can optimize over time the experience that you make available to your consumers. So it's not just product data in itself. Product data is a core part of it, but it's the full of the product experience that we want brand to optimize and to maximize to their customer. Because ultimately, the products don't exist until way past their purchase time in the vast majority of cases.

[00:13:08] So the only way they exist is for the digital experience and connection you have with them. And so this is what's going to drive your conversion rates. This is what's going to drive your return rates. This is what's going to drive the lifetime value of your consumers. And this is what's going to drive your ability to just surf the wave of AI commerce or not.

[00:13:32] And there will be some critics out there that argue that AI is making shopping, yes, more efficient, but also in many ways less personal. So do you think AI can genuinely improve that personalization without losing the emotional connection between brands and their customers? Again, something very important to retailers. So that's an interesting one.

[00:13:56] A couple of months back, the Turing test, so assessing where people can say if a machine is a human or if it's a human, was run on the latest version of ChatGPT. And the outcome of it was not only ChatGPT 5 actually blew through the Turing test, but that people assumed the machine was a human more often than they assumed humans were humans.

[00:14:25] So it means that as individuals, we tend to think that if we don't know who's behind the curtain, ChatGPT is more of a human than actual human. So the ability of the current generation of conversational AI to convey connection to the intent and the people they're talking to is clearly there. So the limit is not really on the technology itself.

[00:14:49] What the limit is going to be is can you have that AI embody what you want your own brand to be? A seller in a shop is different from a brand in another shop, and it's different from that in another shop. And so how do you make sure that the tone of voice, how do you make sure that the expectation, how do you make sure that the behavior of that conversational AI matches the experience you want to give to the end consumer is in my view more of the question.

[00:15:18] So yes, that personal relationship can exist with the technology. It needs to be governed and controlled the right way to match what you want to provide as a brand.

[00:15:30] And I think with giants like Amazon and Walmart setting new benchmarks and leading the way in AI driven retailer, I think the big question for many people listening around the world is how can their smaller or midsize business or retailer, how can they compete with their compete and maintain trust within their audiences too? Well, I think for a small retailer, the question is not whether you can compete with Amazon or Walmart.

[00:15:58] The question is, can you exist in a world in which they exist? And the reply to that is yes, definitely for a number of reasons. First, just on the question of AI and technology and the rest. You don't need to be a hundred billion dollar company to actually use AI technology. It is now available. It's readily available. Anyone can use it. You know, we've got the open AI APIs.

[00:16:25] You've got all of the systems that are there that you don't need to spend hundreds of millions of dollars of R&D to be able to use that technology. The technology is there. Now, the distribution channels, they're there too. Not that long ago, when you wanted to open a new country, that was such a hassle. You needed to find a distribution channel. You needed to translate your information. You needed to do quite a few things today. And distribution channels are there.

[00:16:51] You can actually leverage Amazon or Walmart as distribution channels themselves. And the act of getting ready to sell has been made much easier. For instance, our customers at Akeneo, many of them use Akeneo to actually be able to easily open new markets. Because first, you can easily connect to new distribution channels. But second, you can adjust your product experience to these channels.

[00:17:16] Automatic translation, copywriting, enrichment of your information based on the specificity of the local and the regulation. All of these things used to take months to do and can now be, in large parts, automated. So I do think that smaller brands and organizations should not see AI as a moat for big organizations. I think it's the opposite.

[00:17:40] There were things that only big companies could do as of today, such as open 20 new markets in the next six months, that are now accessible to organizations because so much of it can be automated. And it opens a huge area of the world for better product experiences. You know, it used to be that product experiences was something that could only be done by teams that had significant marketing teams, product teams to be able to work for all of that.

[00:18:07] Now, almost everyone can do it because the barrier to actually be able to operate is much lower. And we're seeing more and more B2B companies, small manufacturers actually jump on the Benvagon. So overall, that transformation of the market is huge. And I don't see that as something that's going to concentrate even more the success. I think that's the great opener that will make way more companies have a chance of success.

[00:18:34] And I do think also, more pragmatically, it will actually have a very positive impact on the cost of acquisition of customers because today, you know, the channels by which you could get customers were very limited in numbers. And with this and these changes, there's going to be more of them. Open AI is only one of them, perplexity. The rest, it's not just Google and Facebook anymore. It's actually a whole gamut of channels that are made available, which will create a bit healthy competition on the cost of acquisition of customers.

[00:19:05] And if we look ahead to life in 2026, what do you think the trusted AI shopping experience of next year will look like in your view? And how do you see customer expectations also continuing to evolve over the next few years? There's an old saying that the last best experience we have anywhere becomes the standard expectation for what we expect to see everywhere. So what do you see changing next year?

[00:19:30] Well, I think, you know, I recently read that 21% of holiday purchases will be AI driven this year. That's up from basically zero last year. I do think that conversational and agentic AI are the big changes that there's nothing we can do against. Agentic a bit further than agentic being the idea that things are being bought on your behalf. I think there's still some work to do before we get there.

[00:19:55] So I don't think trusted AI shopping will be the AI does the purchasing in your seat to begin with. But at least the first step of the AI is working with you to help you identify what best matches the need of what you need at a specific moment. I think this is something that is unavoidable. Everybody will have their own personal shopper where you can share what you want.

[00:20:20] You can share your need of the moment and it will help you and guide you toward what makes the best sense for you at a given moment. I think this, within the next year, will become the standard way of interacting. The same way ChatGPT or Google AI mode are now becoming the standard way of interacting with internet data. Yeah, I've got to ask. I think one thing I've noticed in myself is I'm turning less to Google information.

[00:20:49] I'll go to an AI assistant, whether it's perplexity or ChatGPT. I know my son is very much the same. Do you think retailers and businesses need to be thinking about this and not just appearing on the front page of Google when we're turning away from search engines a little bit? I mean, that should be the biggest question on their mind. It's where are they going to be acquiring customers from in the future? Sure.

[00:21:14] You've got a behemoth that's going to reach a billion monthly active users. Actually, I think they're just getting there. They've launched their buy-in ChatGPT option. I do think that there is a world where there is no more SEO and that search within two years. That this is the end of that model because all of that is a keyword-based model. And the question is, what's going to come next?

[00:21:40] And one potential answer is if you look at ChatGPT that actually is looking to take a cut on transactions, if you take a cut on transactions, you're not relying on an advertising model anymore. You're actually relying on a conversion model. And so the question becomes, how does your product become the highest converting product? And if the question is, how does it become the highest converting product on ChatGPT?

[00:22:06] The only answer is, well, by having the best product data that will go with it. So there is a world that you can think of where the whole of the budget currently being spent on SEO and at search actually becomes budget. That is towards being the best ranked, not on search engines, but on conversational engines. And for that, it's a lot about making sure what is being said about your product and the product data is the right one.

[00:22:34] So in my view, that's a massive thing. And as I said at the beginning, you know, I spent 25 years working in AI. And the main question I always heard is my data is not good enough to do my application. Putting data in the right shape traditionally takes two to three years. The transformation we are seeing is probably going to take 12 to 18 months.

[00:22:55] So companies that have not already started are already late on what is going to be the biggest shift in commerce since, well, at least since the Internet, but probably even more than that. So the one recommendation I can give is look at your product data now. Is it comprehensive? Is it trusted? Is it up to date? And do you know how to make it flow? If you don't, this is the only thing you should be paying attention to.

[00:23:26] Wow. And I think that is an incredibly powerful moment to end on. So much food for thought there. And for anybody listening, having listened to this conversation and sitting around thinking about it, and maybe they want to start a conversation, we will continue this one. Where should they go to find out more information? Anywhere you'd like to point, everyone? I mean, don't hesitate to contact me on LinkedIn if you find a way to spell my name right. That's the only issue.

[00:23:53] But otherwise, no, we've got the akineo.com website, A-K-E-N-E-O dot com. You can definitely reach out to the company there and we'll be more than happy to support you on your journey to provide the best product experience possible and to navigate through the interesting times of the switch towards e-commerce. Interesting times, indeed.

[00:24:17] I will be adding links to everything, including LinkedIn, the website, and some of your other social channels too, so people listening can find out more information there. And I would also add a link to the report around how consumer insights towards AI-powered shopping tools, building trust and rising expectations and so much more in between. But more than anything, thank you for shining a light on this. I, for one, are going to be thinking about this long after we finish recording. But thanks for joining me today. Well, thanks a lot for having me. It was a great pleasure, Neil.

[00:24:47] So AI has already rewritten the rules of online shopping, but as my guests share today, it's just getting started. And the winners of tomorrow won't be the brands that push the most products. They'll be the ones that build the most trusted experiences. And that's going to involve data accuracy, transparency, and authenticity. All these things will decide who consumers listen to and who they just ignore. So what do you think?

[00:25:15] Would you trust an AI to guide your next purchase? Or do you still prefer a human touch when shopping online? Love to hear your thoughts. This is something that impacts each and every one of us. So please join the conversation by visiting techtalksnetwork.com. You'll find over 3,600 interviews there across eight different podcasts. And also connect with me on LinkedIn, X, Instagram, just at Neil C. Hughes. Send me a message. But that's it for today. So thank you for listening as always.

[00:25:45] And I'll speak with you all again tomorrow. Bye for now. Bye for now. Bye for now. Bye for now. Bye for now. Bye for now. Bye for now. Bye for now. Bye for now. Bye for now. Bye for now. Bye for now. Bye for now.