AWS Marketplace: AI, Procurement, and the Future of Partner Solutions – Six Five On The Road

How are artificial intelligence and digital transformation changing the way enterprises approach IT procurement, and what is AWS Marketplace doing to lead that change?

‍

From AWS re:Invent 2025, host Alex Smith is joined by Amazon Web Services's Matt Yanchyshyn, VP, Marketplace & Partner Services, for a conversation on how AI is transforming IT procurement, partner enablement, and the customer journey within the AWS Marketplace.

‍

Key Takeaways Include:

‍

🔹Evolution of IT Procurement: How AI and digital transformation are driving new expectations and standards for efficiency, intelligence, and interconnected services in enterprise buying processes.

‍

🔹Enhanced Discovery and CX: AWS Marketplace now leverages semantic search, clean data integration, and AI-driven product recommendations, making it easier for customers to identify and purchase comprehensive business solutions.

‍

🔹AI-driven Partner Enablement: AWS Marketplace is using AI to streamline partner onboarding, deal velocity, and enable both ISVs and channel partners to deliver more value to end customers.

‍

🔹Agent-to-agent Innovations: The future of procurement is moving towards AI-powered agents conducting transactions, providing smarter, faster, and more transparent experiences for businesses.

‍

Learn more at Amazon Web Services.

‍

Watch the full video at sixfivemedia.com, and be sure to subscribe to our YouTube channel, so you never miss an episode.

‍

Or listen to the audio here:

Disclaimer: Six Five On The Road is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded, and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors, and we ask that you do not treat us as such.

‍

Alex Smith:

Six Five here at AWS reInvent and I'm delighted to be joined by Matt Yanchyshyn, who is a VP of AWS Marketplace and Partner Services. Matt, it's been a very busy week, lots of exciting announcements. Thank you first for taking the time to hop on and talk with us here today. Thanks for having me. So we all know that AI is transforming really every business process and that really includes procurement. Tell us a little bit about how you're seeing IT procurement evolve and how AWS Marketplace is really evolving to meet those needs.

‍

Matt Yanchyshyn:

Yeah, I mean in a lot of ways procurement itself has stayed pretty similar. It's still sort of going through all the phases, discovery, legal, finance, depending on the size of the company and what you're buying. You need to look at multiple products. You need to make sure you're getting the best price, the best deal, the functions you need. What is changing is how fast customers are trying to do each of those stages. And so that's where we're applying AI as sort of all parts of the buyer journey, whether that's discovery or evaluation or deployment. And that's really been our focus speed.

‍

Alex Smith:

And so tell us about some of the announcements that you had around meeting that speed and helping customers discover and purchase solutions on the marketplace.

‍

Matt Yanchyshyn:

Yeah, so AWS Marketplace has about 30,000 listings from over 6,000 vendors today. So it's a lot to work through. And so first of all, we launched a new search experience that's powered by AI, and it does things like an AI-powered summary at the top and it uses AI models like Cohere Rerank and Bedrock to rank the rankings or sorry the results better in the search results. I think what everyone is talking about this week though is agent mode which is another feature that we launched for Marketplace and that's more of a conversational interface powered by AI that walks customers through, you know, they might not know what they're looking for, like they're looking for a zero-trust solution or an intelligent document processing solution. And then agent mode actually uses AI to auto-generate personalized product comparisons for things that correspond to that use case. So that way you don't need to say, hey, I'm looking for a Trend Microvision 1, or I'm looking for Zscale or ZPA. You can say, I'm looking for a zero trust solution and then it allows you to compare those and other products and then actually customize those comparisons. So that's new and the ability to upload whole RFPs and tech specs and have us parse that for you and make recommendations instead of just one line searches. It's a lot more powerful and we're using a ton of AI not just to rank search results but to actually intelligently process your information and help you compare options to make sure you're getting the best product in your needs.

‍

Alex Smith:

Yeah, the one thing I love about this age of AI that we're in is just it really democratizes technology for non-engineers like myself. This is actually probably one of the first times in 20 years in tech that I could sit, watch your keynote, you doing a demo on keynote, and then be on my laptop doing the same thing and seeing great results from some of the early experimentation that I had. Hopefully customers are kind of out there trying it the same. Yeah, I'm glad you're doing that.

‍

Matt Yanchyshyn:

That's what we want customers to do is to honestly stop listening to me and pick up their phone or their laptop and start running some searches themselves.

‍

Alex Smith:

Yeah, absolutely. But you also talked about all that is around the search, the discoverability. Velocity is right through to kind of close one. And especially in the world that we're in, sales is always about fast, fast, fast. Are you doing anything post discoverability and search to help customers procure faster?

‍

Matt Yanchyshyn:

Yeah, I mean, you're totally right. That's just the first phase. And I don't even think it stops at close one. You know, there's also what happens after you buy. And we're really looking at all of the phases of the buyer journey. So agent mode, the Gen AI search for discovery, but also for the transaction piece, we launched something called Express Private Offers. And what that does is really shorten one of the longest times typically in the sales cycle, which is the negotiation. You know, most customers, well, actually a majority of customers these days buy through self-service. Over 80% of our transactions are self-service transactions.

‍

Alex Smith:

And self-service means there's no... Yeah, pay as you go.

‍

Matt Yanchyshyn:

I know what I want, or I do a free trial, and I subscribe on the spot. Like Amazon.com, click and go. But the bigger transactions, and certainly enterprise transactions, there's often a formal RFP, RRQ process, and they get discounts. They say, hey, what if I buy a contract for two or three years, and I'm a certain type of company, I'm good for it, will you give me a discount? Most companies have a rate card. We call it private pricing or private offers. So what we launched is private pricing as a service or express private offers. And what that allows you to do is you enter some information about your company, but also how long do you want your contract for? How much are you buying? The things that normally salespeople would collect and then negotiate. But instead of that phone call, email, It's not a manual negotiation that can take weeks. We essentially say, OK, are you who you say you are? We qualify the customer based on the company information. And then do you meet XYZ criteria that the seller sets up? And then we issue that private price with the discount on behalf of the seller. And that's huge. Obviously, it's not going to be for your massive $100 million deals, but maybe a $10,000, $50,000, $100,000 deal, where it's a two-year maybe renewal, you kind of know who the customer is. You don't necessarily need a massive sales team on that. So we can automate that private pricing automation and shorten it to minutes or seconds. And that is a huge deal, because it's this huge middle space between fully self-service and fully custom private offer. And we think customers will really love that AI-powered automation. And the AI-powered part of that is that customer qualification.

‍

Alex Smith:

Yeah, and presumably the seller on the other side can tweak their conditions based on their criteria, their potential time of year. Yeah, exactly.

‍

Matt Yanchyshyn:

They're in total control. They have a rate card, they have a discount sheet, and they basically take those exact same rules and they load it into Express Private Offers.

‍

Alex Smith:

Yeah, it really is just kind of from discoverability to, you know, valuation to procurement. It's just, you know, increasingly digitizing through intelligence as well, the whole buyer journey. And it's amazing to see how, you know, just the evolutions at kind of all stages of that buyer journey. Now, beyond the actual procurement side of things, how else are you helping customers really embrace AI within their own operations? I mean, I know you've launched new categories and segments around AI. You have your own protocols and everything for helping integrate agents within their operations. So could you talk about some of the things you're doing to make it easier for customers to just kind of digest AI more broadly into their organizations?

‍

Matt Yanchyshyn:

Yeah, well, I mean, first of all, some of the features I just mentioned, like Agent Mode, you know, you were saying you were trying it on your laptop. I've been showing everyone on my phone, if you open up Anthropic Cloud, like the app, and you have the pro version, there's actually a connector marketplace, a connector catalog they have, and there's a new AWS Marketplace connector. And that's actually our Agent Mode feature with MCP. So it's an MCP server that you can then use to run those same searches that you would on Marketplace through an AI-powered tool. And that's really powerful because then you can, not just on your laptop, but on your phone, using the AI tool of your choice via MCP, make those product comparisons. And it's the same thing, same exact information. You get the reviews, you get the summarizations, you get the product comparisons. So that's sort of one way we're extending the features like some of the ones I just described into tools that customers are using every day. Increasingly, those are AI tools in your pocket. So another big thing is the AI agents themselves and energetic solutions. So we launched an AI agents and tools marketplace back in July at the New York summit. We talked about it. And that has grown substantially. So we originally started with a target of 50 listings in that category of the marketplace. And I told my team, I always like to say, add a zero. And they added a zero and then some. At launch in July, we had over 800. AI agents and tools in the new category, and that's since grown to over 2,000. So I think that's amazing growth, but honestly, that's not as much what matters. We can have the biggest catalog in the world. What matters is that customers are increasingly consuming them. We're seeing a really large number of both self-service and private offers for actual agents. I think we were just saying that it doesn't stop at discovery. What makes this special is not just having the agents on Marketplace, but we make it really easy to then deploy those agents into Amazon Bedrock Agent Core, so you can start to use those agents right away. Because Gartner, for example, has a ton of research that up to 25% of all SaaS applications rarely, if ever, get used. I think it's pretty true based on what I see and the same can be true for agents. The more easy we can make it to find the agent you need in the marketplace and then deploy that agent to Bedrock Runtime and Gateway so you can start using that agent, the happier the customer is. So again, that same buyer journey applies to the new agent category.

‍

Alex Smith:

I think going back to discoverability, I think there was a time where you could almost think that just the degree to which the catalog is growing could just become too much of a minefield. But I guess now with AI tools and just different ways of discoverability based on customer needs, it almost doesn't matter how many is out there, right? Because AI can really sort through and find it to your very specific needs.

‍

Matt Yanchyshyn:

Oh, you're totally right. We spent a lot of time previously thinking about how much is enough. What kind of size catalog should we have? How many vendors? You're right. With the new AI-powered search and agent mode, it doesn't matter as much because these are powerful AI tools. We use Anthropic Cloud, Emission Coherent Rerank, we use Nova, and they can really quickly parse through huge amounts of information and help you zero into the specific product that best meets your needs.

‍

Alex Smith:

Yeah, it's almost now probably more is better.

‍

Matt Yanchyshyn:

Yeah, I mean, Amazon.com has known that for years. And that's why they have powerful search and AI tools like Rufus. And we're applying a B2B context to that for software and professional services buying on the marketplace.

‍

Alex Smith:

But no name like Rufus for your report?

‍

Matt Yanchyshyn:

Well, if you saw the keynote yesterday, you would have seen I was wearing a dog, a cartoon dog. So maybe it was a bit of a shout out to Rufus. But no, we're serious enterprise.

‍

Alex Smith:

We've talked a lot so far about the customer journey, but I think the other important side of obviously what you do and making it success for those customers is to have a good experience for all your partners that are on the marketplace and providing software and services. So have there been any kind of if you put your innovation engine to kind of help some of that side of the camp?

‍

Matt Yanchyshyn:

Yeah, first of all, I mean, my title is actually AWS Marketplace and Partner Services. So I wear two hats. I spend a lot of time focusing on the customer experience, how they discover partners and solutions they need, but also the partners themselves. So first of all, just starting with the marketplace, we launched something that also definitely benefits customers, but partners are really excited about, called Multiproduct Solution. And that, if you look at the marketplace, we were just talking about listings. There are about 30,000 listings. 10,000 of those are actually professional services listings, people services. And those professional services providers like Presidio or Accenture or Deloitte, they want to more effectively sell their services along software. And many of them are selling software now. And so we give them the ability to package up software and services from multiple vendors in a multi-product solution listing that they can then sell and go to market with us to customers. So that's really powerful because a huge benefit of working with AWS is our go-to-market, helping you co-sell. And it's easier to co-sell with us if you have a customer use case specific multi-product solution that combines professional services from our partners with software from one or more vendors. That gives us a solution-based sale mechanism, and all of our partners want solution-based co-selling. So that's definitely something we launched with Multiproduct Solutions and something we're focused on. On the more back-end for partners, it's just like we were talking about making procurement faster for customers is so important. The same for partners. If they can't make their operations, like their cost of sale low and be efficient, then they're going to have lower margins and their operating costs are going to be higher. So we launched Partner Central 3.0 in the console. We moved our main interface that partners use to do business with customers into the AWS Marketplace console. And in and of itself, it's like, that's just a move. But with that move, we launched a bunch of new APIs and also more sophisticated access control So for example, if you're an alliance manager that does something really specific, you can get access to do that specific thing. So you can then think of the personas in your partner organization and assign them really specific duties and give them permission to do those things either on the partner central website or programmatically via our new APIs, like a benefits API. You can go and request those credits without picking up the phone and calling someone on the programs team. So it allows them to move faster, lower their cost of operations, integrate their partner tooling into their CRMs and ERPs and other systems that they're using. So the same applies to partners in customers through a mix of AI agents, API integrations, more sophisticated access management, they can operate more efficiently. We want to be the best place for them to grow their business, but also the most efficient place to operate.

‍

Alex Smith:

I always say the buyer and the seller side, it's two sides of the same coin. You can't have everything you're trying to achieve on customer velocity side of things, you can't achieve if you're not putting those same kind of innovations and mechanisms for your sellers to be able to meet the customers at the speed that they want.

‍

Matt Yanchyshyn:

Yeah, we have hundreds of thousands of customers and 3 million plus subscribers on the marketplace. At that scale, to meet that customer demand, you need to be able to operate really efficiently. And your cost of sales can't be high. You need to have good operations. And to do that these days, you need to use AI, you need to use API-based integrations. Otherwise, you won't be able to capitalize on all of the customer needs and all the business out there.

‍

Alex Smith:

Yeah. One thing as well you've kind of touched on, and I should come back to it on your announcement with the Cloud Anthropic. I almost see this as kind of an extension or a different lens of the buy with AWS that you, I think, announced last year. Yeah. It's really increasing the surface plane of where marketplace is, right? Now it's not just about coming onto, you know, awsmarketplace.com, it's making sure that you can get access to and kind of benefit from all that kind of discoverability and power wherever you want to kind of start your buyer journey, right?

‍

Matt Yanchyshyn:

Yeah, you're 100% right. We call it meeting customers where they are. And we were focusing, and we continue to focus on AWS services. Like if you get Oracle DB through RDS, or if you buy a third party, a partner add-on through our Kubernetes service EKS. Or we just launched partner integrations in our network firewall service. There are a number of AWS services where we extend Marketplace listings to be surfaced there. Because if you're a Kubernetes administrator, you're in the Kubernetes console, our EKS console, and that's where we want to put the Marketplace listings. So we have this concept of embedded listings and mini-Marketplaces, like Bedrock Marketplace and all these other Marketplaces that are powered. by the marketplace. But what we did with AWS is extend that beyond the walls of AWS. So now partners like Databricks can have their marketplace listing on their website or in their application because that's where their customers show up. And they're saying that the conversion rates on those, you know, they go from free trial to marketplace purchase without ever leaving their application, are way higher. So that's really proven out as a model and we're taking it to the next step. We just announced an acquisition of Phoenix.ai and that's a storefronts company. They do a lot of things for procurement but one of the great things that they've done is build procurement portals really in storefronts for third-party partners. And that will allow us to further embed marketplaces and whole procurement purchasing experiences into enterprise procurement portals or channel partner procurement portals. So I'm really excited about, you know, especially exposing our partners in more places so they can reach more customers and do more business. Because SMB customers, line of business buyers, they all have, they all show up in different places and we want to be everywhere they are.

‍

Alex Smith:

Well, it's a busy week on top of all the announcements, announcing an acquisition at the start of the week.

‍

Matt Yanchyshyn:

Yeah, we were high-fiving each other last night. It's very exciting.

‍

Alex Smith:

All right, so I guess getting close to wrapping, I think this is an unfair question given just how much you guys have been sprinting to kind of get all these latest innovations out and all the releases out this week at reInvent. Now kind of like looking forward, where are you pointing your compass kind of from an innovation standpoint? I mean, you know, just directionally, what are some of the things that you're kind of, you know, you and your team are prioritizing in terms of, you know, continuing on this kind of AI journey within the marketplace?

‍

Matt Yanchyshyn:

Yeah, well, first of all, I'm no different from my customers and my partners in the sense that my own team needs to also, you said sprinting, we're sprinting much more efficiently with AI ourselves. We launched 27% more features year to date, and we're about 31%, if you normalize it, like how we measure developer productivity, software development productivity, 31% more productive year to date as well. So we're shipping more features, more code, to keep up with the demand, because we expect multi-product solutions, express private offers, agent mode. to continue to do well. And the reaction has been great. And with reaction comes many feature requests. And we're going to have to keep up with those feature requests. So I think continuing to focus on how we ourselves, my engineering team as a customer, can continue to use the same AI tools, Bedrock and Agent Core, to be efficient ourselves so we can keep up with this growing demand. That's a top priority. I spend a lot of time thinking about how can we ourselves move faster so we can keep up with those needs. And I think it's really doubling down on what we started. We're just getting started with AI and automation when it comes to the partner journey. You should expect to see a lot of agentic capabilities in partner central down the pike. And for buyers, express private offers is just the beginning, as is agent mode. We're doing product comparisons with first and third party data powered by AI. We're doing pricing as a service. But there are a lot of other parts of the buyer journey. There are legal parts. There are other finance parts. There are governance, risk, and compliance parts. that I know I can make faster and I know that as AI continues to improve these things that were previously unimaginable, you know, touching those parts of the procurement journey and really compressing the time are possible now. So we're really, I know it's kind of cliche in the Amazon sense, but it is truly day one for us and the buyer journey is large and complex and there are a lot of parts that we'll continue to push on.

‍

Alex Smith:

I really saw in your eyes there when you kind of highlighted all those different functions and talked about trying to make them faster. So no doubt I could see the innovation kind of... I was projecting to the next re-invent already. Yeah, absolutely. So Matt, thank you so much again for taking the time to speak with us. I know it's been a very busy week for you here at re-invent. Congrats to you and the team for all the announcements. I look forward to catching up again in the future. And to all of you that are tuning in, thank you so much for taking the time and stay tuned. We're also speaking with some of the Marketplace partners shortly to hear about their experience and journey on the AWS Marketplace.

‍