Elevate — The Global Commerce Summit 2025 | Shiri Mosenzon Erez, commercetools

Clips
More from Elevate — The Global Commerce Summit 2025

Shiri Mosenzon Erez

Chief Product Officer

commercetools

play_circle_outline Revolutionizing Commerce: AI Hub Announcement Marks Start of New Era in MCP Evolution

play_circle_outline AI-driven conversational commerce and fully generated experiences

play_circle_outline Hyper-personalization for end consumers and B2B buyers

Info
Transcript

Shiri Mosenzon Erez, commercetools

Shiri Mosenzon Erez

Chief Product Officer commercetools

Shiri Mosenzon Erez, chief product officer of commercetools GmbH, joins theCUBE’s Scott Hebner and Jennifer Wu at Elevate — The Global Commerce Summit, for a forward-looking conversation on AI and the future of digital commerce. The discussion focuses on how emerging technologies are reshaping customer engagement and infrastructure.

Mosenzon Erez highlights commercetools' AI Hub and Commerce MCP as foundational tools for enabling dynamic, personalized shopping experiences. From conversational commerce to integrations with technologies such as Meta sm... Read more

explore Keep Exploring

What was the announcement made by commercetools and what is the purpose of their AI Hub and Commerce MCP? add

What is the evolution of shopping journeys and personalized advertisements? add

What will the shopping experience be like for everyday consumers in the near future? add

bolt Powered by CUBE AI

Shiri Mosenzon Erez, commercetools

search

Patricia Wu

>> Hello and welcome to Elevate, the Global Commerce Summit. I'm Patricia Wu, your host, along with my co-host and analyst, Scott Hebner. So Scott, you are going to be a kid on Christmas morning this next interview with your passion for AI.

Scott Hebner

>> That's right. Yes.

Patricia Wu

>> All right, and no further ado, we're going to introduce our next guest, Shiri Mosenzon Erez, chief product officer at commercetools.

Shiri Mosenzon Erez

>> Thank you very much. It's great to be here.

Patricia Wu

>> Oh, it is so great to have you here, especially after that cool announcement and that product demo this morning. Tell us about it so that our viewers who didn't get to watch it in person can relive it with us.

Shiri Mosenzon Erez

>> Absolutely. I think just taking a step back, at commercetools, what we're trying to make sure we set up for is the future of commerce as a whole. And with the AI revolution, which is where we are at the moment, after we had an age of a monolith from a commerce perspective, moved on to headless and composable, it's now an entire new age, and it's really changing how customers in the next few years are going to be conducting their shopping journey. So, the announcement was around our AI Hub, and our Commerce MCP and really those two things, they're there to enable this futuristic, which is really rushing upon us, experience for customers that are basically shopping. So I mean, I can tell you a little bit more about how I see that journey and then tell you a little bit more about the Commerce MCP?

Patricia Wu

>> Yes, please do. All of it.

Shiri Mosenzon Erez

>> Okay, great. So, I think we touched upon it in the session with Dirk over there. What we did was we had Meta glasses, because wearables, and having things on you as you shop is something that is also becoming more and more easy to do, and therefore is prevalent. And Dirk, through his glasses, was able to say, "Oh, I would like to buy this jacket specifically," or a wristband that I had on, and taking a picture of that, you can already start a shopping journey. It can happen in the brand real estate. So identifying the brand and then going onto the branded site of the product that you scanned with your Meta glasses in that example. Or it could continue the journey through an AI chat line style thing. For example, like ChatGPT, or Microsoft Copilot or Gemini, any of those. Those are two commerce journeys, and what we did on stage was we actually showed how this can connect to one of our customers' catalogs, and you can complete the entire shopping journey that started with a picture of what you wanted by just dialoguing with the AI.

Patricia Wu

>> Wow, that is so amazing.

Scott Hebner

>> Yeah.

Patricia Wu

>> I mean, I think about the number of times I've seen something like maybe on TV, something that somebody's wearing, and I say, "Where did she get that gorgeous blouse?" And that's here now.

Shiri Mosenzon Erez

>> Exactly. And it's not just about... So, there is the fact of identifying the object that you're trying to buy, and being able to understand that, but the entire journey of the dialogue that you have becomes fully generated, and that's what we saw today. So in the past we had journeys, shopping journeys that were quite static. You'd go onto a website, you'd see what you see, this is what's on offer, you scroll down, you make your selection. Then it moved to be more dynamic. So, if they know that you love the color blue, as an example, then you'd get advertisement for things that are blue, or based on what you previously shopped. That's more dynamic. In the future, and what we saw there, the entire screen is generated at every given moment, and it's responding to where you are at that given moment, what you've bought before, what you're asking, what you're dialoguing with the AI about. And it's not just the products that are shown that are different. Even the headlines were responding to what Dirk was saying over there on stage.

Scott Hebner

>> Yeah. What really jumped out to me was it wasn't a transactional experience, it was more of an emotional... Because you're going through a little journey on what you're trying to purchase. I thought the demo was really, really effective.

Shiri Mosenzon Erez

>> You don't even have to touch the screen. Right?

Scott Hebner

>> Right.

Shiri Mosenzon Erez

>> It's all by voice. It's all conversational as well. So there's a few elements here, and all of them are using different AI models in order to be able to exist. Right? There's the image recognition, which is AI as well. There's the LLM element of understanding, what are you saying? You're saying you don't want a black... You don't want a blue wristband, you want a black one. Understanding that you want to now talk about the address, or the shipping method, that's conversational. And then this is the real agentic piece is actually actioning that order on your behalf. That's agents, that's agency.

Scott Hebner

>> So give us a little bit of background on the agentic toolkit, and the AI Hub that you announced, and then we'll talk a little bit about MCP.

Shiri Mosenzon Erez

>> Excellent.

Scott Hebner

>> Then we'll bring it back to how it's being applied.

Shiri Mosenzon Erez

>> Yeah, happy to do that. So, the AI hub is around... There's mainly two things. So one is it's being able to use any external AI tools that you wish, because a lot of these are being developed, the big tech companies are creating a lot of these, and our retailers, or our brands might want to connect them into their journeys, so that's one thing. And the second thing from an AI Hub perspective is this what I call, I guess a good way of describing it is, enabling a brand to have the buy button when a customer is doing conversational commerce that doesn't start with the brand. So it's dubbed chat line commerce, I think, sometimes. The names are still shaping, right? But, this is about, okay, if I asked for an item, I'm talking to my AI assistant that is based on one of these big tech companies, will I be able to complete the purchase there? Or is it going to end up sending me into the brand's site? And this is a big dilemma for people here at Elevate. Do I want to bring the customer into my real estate, into my site? Or do I allow this ultra convenience piece and to complete the journey there? The AI Hub allows you to have the buy button no matter which of the big AI, big tech companies you're starting your shopping journey with.

Scott Hebner

>> So basically any agent or AI model that's enabled by any of the big tech companies-

Shiri Mosenzon Erez

>> We're going to-

Scott Hebner

>> Can then... Yeah.

Shiri Mosenzon Erez

>> Exactly.

Scott Hebner

>> Then it can be plugged in. And this sort of builds on where I think you guys really set yourself up nicely for the future, is the MACH architecture, right? MACH right?

Shiri Mosenzon Erez

>> Yeah. MACH. Yeah.

Scott Hebner

>> MACH.

Shiri Mosenzon Erez

>> I never know how to say it.

Scott Hebner

>> Microservices.

Shiri Mosenzon Erez

>> Exactly.

Scott Hebner

>> It's API based.

Shiri Mosenzon Erez

>> That's right. Headless.

Scott Hebner

>> It's a cloud , and it's headless.

Shiri Mosenzon Erez

>> Yeah.

Scott Hebner

>> Which equates to the ability to evolve quickly with new technologies, plug and play. I mean it's the whole composability thing, right?

Shiri Mosenzon Erez

>> Exactly.

Scott Hebner

>> So I think what you're saying now is you're going to be able to enable retailers to do commerce at the point of wherever someone's engaging, particularly if it's through AI that they didn't even go to your website. Right?

Shiri Mosenzon Erez

>> Exactly.

Scott Hebner

>> It's right . Yeah.

Shiri Mosenzon Erez

>> It's meeting the customer where they want to shop. And that's the ultra convenience, the ultimate convenience.

Scott Hebner

>> And that's where the MCP comes in?

Shiri Mosenzon Erez

>> Exactly. So we talked about, we announced our commerce MCP, which is basically a layer that allows... It's a layer that includes some tools, tools for agents, but also an MCP protocol that basically explains to an external agent what is available for them to do. So this is... You know, MCP is... That was a standard that came from Anthropic, and it's really about describing to the AI agent, which understands natural language, so it has the LLM elements in it, right? Every agent has an LLM element in the ins, and also in what it can produce. The MCP, the commerce MCP allows it to do whatever is needed, and to use its judgment almost, within the commerce tools environment. So, you could ask an agent to create an order, you could ask an agent to update a product catalog. You could have an agent that looks at prices elsewhere, and updates your pricing accordingly. It's limitless, in terms of the flexibility, and it is all enabled by that MACH architecture underneath, and the APIs.

Patricia Wu

>> So we've talked about convenience, we've talked about flexibility. Are there other benefits for your customers, and the end consumer?

Shiri Mosenzon Erez

>> I think the key thing is, like you said, the end consumer. And this works for B2C customers like you and I, when we're shopping for something, but also for B2B. It's about commerce journeys, and we want to make those seamless. We want to make them personalized, as they need to be. I think both end consumers, and also buyers in B2B, they have expectations for things to just kind of work, and for systems to just understand them. And up to this date, we were kind of there, but now it's really coming, and it's really going to be what we enable through this Commerce MCP, and also the AI Hub.

Scott Hebner

>> So you jump out a year from now or so, for the everyday consumer that's out there shopping, how is this going to be different in terms of the actual experience?

Shiri Mosenzon Erez

>> So, this is going to be significant. First of all, hyper-personalization, for sure. So, it'll know if you walk into a store, or if you start your journey with your wearable device, whether it's the glasses like Dirk had, or maybe some wearable watch, it'll know everything it needs to know about you as a customer. It'll also know everything that it needs to know about the product if you're asking about a product. So, that unified lens is an important backbone for this, all sitting on really great data, which becomes ever more important in the world today. And then the experience that you get is really hyper-personalized. So, advertising will be accurate. What you get on screen, and what isn't shown on screen is all going to be taken into account to help maximize conversion. So, it'll be really convenient for you, if you're looking for something to find the right thing more than ever. And it'll be really easier for brands, and retailers to get you, right? To help you find what you want, and convert, and check out seamlessly. So, an extremely personalized, really convenient journey that is tailored exactly to what you need.

Scott Hebner

>> And the personalization, as I understand, it goes beyond the ads being tailored and all that, but it actually can create a workflow that's unique to the goal you have. So you don't have to be as explicit with the prompt. You can say, "I want to buy a new watch."

Shiri Mosenzon Erez

>> Yeah.

Scott Hebner

>> And then it will create a workflow among the appropriate agents under the covers. So it's a whole experience, right?

Shiri Mosenzon Erez

>> Exactly.

Scott Hebner

>> Yeah.

Shiri Mosenzon Erez

>> And this is already happening. If you go on one of these chat lines, whichever one, it will respond to a prompt that says, "Oh, I'm looking for a new watch." It'll show you something. What we want to make sure we enable with Commerce MCP, and with AI Hub is that for brands and retailers, they can also have this kind of experience when it's inside their real estate. So, inside either their physical store, but also on their website, or on their app. So, it'll be everywhere. And actually what we didn't touch upon, it's great to be focused on the end customer. I think that's definitely number one. But what we didn't touch upon is that this is also really... This technology is going to be really useful for merchants, for shopkeepers, when you try to curate your catalog, or manage anything, advertising, or any content that needs to be managed, you can, through the Commerce MCP, you can make changes and your workflow as a merchant becomes easier.

Patricia Wu

>> Wow. I keep thinking about all the time saved with the things that are not showing up.

Shiri Mosenzon Erez

>> Exactly.

Patricia Wu

>> I can finally write that great American novel.

Shiri Mosenzon Erez

>> Oh yes.

Scott Hebner

>> Yeah, I mean-

Shiri Mosenzon Erez

>> I look forward to reading that.

Patricia Wu

>> Well, speaking of dreaming, I'm wondering, beyond the immediate use cases, are there any unexpected areas where you think this agentic toolkit and AI Hub can unlock new possibilities?

Shiri Mosenzon Erez

>> I think that's really interesting. I think that goes back to the fully generated experiences, and where they're actually going to show up. Like how are we going to... What does front end look like in the future? Do we look at our phones anymore at all? Or do we really move into a fully conversational dialogue, which then you've got something that the screen comes up like in Iron Man, right? When you have this 3D thing. I think that those are yet to be clear on how the interaction is perfected, right? So yes, it'll be fully generated, but kind of where, I feel, is something that is going to be unexpected, and I look forward to seeing what is... As a person, even for me, I'm thinking, what is the most convenient way for me to interact? I'm not sure that I know, and now all the options start opening up. So, that will be interesting as well.

Scott Hebner

>> That's the power of having an open platform like you have, is you're just going to unleash innovation.

Shiri Mosenzon Erez

>> Exactly.

Scott Hebner

>> In addition to what you just talked about, I think it's also just the movement from the transactional experience to an actual personalized experience, I mean, it becomes an experience to shop. Right? I also like the idea that it's not just responding to you, but it's anticipating.

Shiri Mosenzon Erez

>> Yes. Anticipating.

Scott Hebner

>> It kind of figures out, and helps you, guide you on what you want to actually accomplish or do, and come up with new ideas. And you can imagine how the whole experience is just going to be radically different.

Shiri Mosenzon Erez

>> Yeah. It's going to be exciting. And I think it's also important to mention that underpinning that, the things that we need to be thinking about, and others need to be thinking about are the memory of these agents, and how much of that to have, so that you can really leverage. But on the other hand, there's a cost element to it. And then secondly, and this should probably be first, is the security. So, if we have all of these agents, and lots of companies are going to be building great agents, making sure that what you allow to reach into your system, and to write to your system is fully secure, and fully compliant, and all of that, and that's another thing that we are taking great care of.

Scott Hebner

>> Yeah. Trust is the currency of innovation.

Shiri Mosenzon Erez

>> Trust. Exactly.

Scott Hebner

>> Yeah. And it comes from governance and security, and if they don't trust it, they're not going to use it. Right?

Shiri Mosenzon Erez

>> Exactly.

Scott Hebner

>> So, especially when you're shopping, you're spending real money.

Patricia Wu

>> That's true.

Shiri Mosenzon Erez

>> Exactly. There's transactions in play.

Scott Hebner

>> Yeah.

Patricia Wu

>> So it sounds like the future is seamless, it's hyper-personalized, it's connected. I'm looking forward to it.

Shiri Mosenzon Erez

>> Exactly. We're in it. It's happening.

Scott Hebner

>> Yeah.

Patricia Wu

>> It's now. The future is now.

Shiri Mosenzon Erez

>> The future is now. It's great.

Patricia Wu

>> Thank you so much, Shiri. This was so much fun.

Shiri Mosenzon Erez

>> Thank you. It was great to be here.

Scott Hebner

>> Thank you.

Shiri Mosenzon Erez

>> Really interesting conversation.

Patricia Wu

>> That was great. Thanks so much.

Shiri Mosenzon Erez

>> Thank you.

Patricia Wu

>> Well, thank you for watching our coverage of Elevate, the Global Commerce Summit. I'm Patricia Wu for Scott Hebner, and of course we're going to have a lot more coming your way. You're watching theCUBE, the leader in enterprise tech news and analysis.