Anyone who’s ever been to CES will tell you, it’s a biiiggg trade show. While a lot of the action happens on that vast trade show floor — where everything from autonomous cars to massage chairs are demoed — Melissa Burdick, Co-founder and President of commerce acceleration platform Pacvue, said there is a whole other hive of activity in the suites of the adjacent Aria hotel. (Get your mind of out the gutter, we’re talking business here.)

In this alternate CES, executives spend their days meeting, connecting and discussing the future of their business with topics ranging from AI (of course) to retail media challenges and opportunities to the expansion of connected TV (CTV) in Burdick’s case. “I spent all my time in the Aria suites; you could have launched some flying saucer and those of us in the Aria suites would never know,” said Burdick.
Burdick took Retail TouchPoints behind the scenes of her meetings with all the top players in retail media, from Amazon Ads, Walmart Connect and Instacart to Group M, 84.51° and Target’s Roundel to share what is top-of-mind for retail executives in 2025.
Retail TouchPoints: What topics were really driving conversations at CES?
Advertisement
Melissa Burdick: For us, retail media networks were the topic. There are 253 of them now — and counting. And what I thought was an interesting perspective from Samantha Bukowski [Global Head of Commerce] at GroupM was this concept of “head” and “tail” retail media networks. The way the industry is going is, you’ll buy the head — which is going to be Amazon, Walmart, Instacart, Target and Kroger — you’ll plan and optimize and activate those [individually].
But then with this long tail of everybody else, you’re going to want to programmatically manage it, so you put money in, have it optimized and get results based on your goal, whether it’s traffic, incrementality, ROAS, etc. Because as a brand, you’re not going to buy 253 media networks. There may be a lot of networks that you want to buy, but you don’t want to buy them one by one.
The whole process of planning, optimizing, activating and measuring all these retail media networks is very cumbersome, so that was another topic — how do we make these workflows easier? This goes back to the broader AI conversation.
I feel like the conversations around AI right now are very nebulous and high level. There are some obvious things like copilots, but there is a lot that is still TBD. At the Amazon Advisory Council, they had a speaker who is an AI expert and puts on these workshops, and he shared how he helps companies get more tactical in their application of AI.
You get into your functional teams — finance and marketing or whatever — and you talk about all the things you want to automate. Then you go research what new tools [can help you do that] and make your life easier. There’s a new tool every day that’s popping up; for example, ChatGPT just launched Operator. So, you allocate half a day each week for three months to [testing these AI tools] as a team.
This way instead of all these vendors coming to you saying, “Use my AI tool for this; use my AI tool for that,” the way to think about it as a brand is, here are my tasks that I want to automate. What tool can I use to do that? Everyone’s talking about AI and they’re saying it’s coming, which we all know, but the actual application and what tools to use for it is the thing to figure out in 2025.
The other thing that I’m seeing at the same time is middle management is getting laid off. There was a bunch of middle management sales staff laid off at Instacart, Facebook, Amazon, and Google said [it’s preparing for lay offs as well]. With AI, you need more doers. For example, Salesforce is getting so much better with the AI tools they’re integrating that provide better reporting and measurement of the teams. You may be able to cut out some of the middle management because the tools are getting better at doing some of that work.
One other big theme was the democratization of CTV so that everyone can buy it. It’s not just through an agency where you need a kajillion dollars to buy anymore. But it’s also about democratizing the creative, and that’s where generative creative comes into play. The more creative I can create and cut up into the different sizes that I need, the more advertising I can do.
RTP: Were there any other recurring retail media themes that came up?
Burdick: Full-funnel marketing — retailers want you to use all their stuff from upper funnel to lower funnel within their walled garden tech stack. It makes a lot of sense if you look at the Amazon playbook. Amazon has so many offerings, and it’s all around their concept of full-funnel marketing. As retailers look to increase their [retail media] revenues, they’re increasing their offerings.
Incrementality also was a huge theme, and that goes back to the fact that there are 253 retail media networks, but people don’t have bigger budgets. If they are going to spend more money, it has to be incremental [to their brand]. Brands are asking, “Why should I spend on XYZ media? Am I getting a new customer?”
A good example is Thrive Market launching with Instacart Ads. Thrive Market is very niche; they have a very specific audience, so it’s a different way to hyper-target and get new customers by placing ads on Thrive Market. It’s like GoPuff, which is an example from a long time ago. They had a stat that the majority of purchases [on their site] happen between approximately midnight and 5am. That really resonated with people because that’s a different customer than the person who’s going to Walmart during the day.
What brands want is more market share, that is, more customers, especially since brands are really struggling today. Costs are increasing, and their market shares are shrinking. The only way for regional media networks to work is by attracting brands to spend on them, so they have to prove this incrementality.
This premise drove the questions that we were asking retailers: What’s unique about your network? What’s your hook? How are you attracting customers? Sam’s Club, for example, was talking about the unique experience in their Clubhouses and how they have this scan-and-go technology — that’s one of their unique value props.
RTP: I just saw a column from some analysts at Forrester who were saying that everybody looks at Amazon and thinks they’ll have that same level of success if they launch their own media network, but they’re actually not going to be able to. And as you say, new RMNs keep coming, but brands only have so much to spend.
Burdick: One of the things about Amazon is they’re fast, and retailers are just not that fast. They’re not tech companies. Sephora is a really good example. Their merchants are so worried about the customer experience of what shows up on the page that they want complete control, so everything is a managed service on Sephora’s media network. They haven’t gone to self-serve services, which is how you scale a media network. Conversely, Amazon launched CTV in two years. It’s tough to compete with that speed.
Also, Amazon has a totally different cost model with AWS [Amazon Web Services, it’s cloud division]. Everyone’s building these data clean rooms to compete with Amazon Marketing Cloud, but those have a cost; they have to pay someone like Snowflake or Habu. There’s a different cost structure [for retailers that] don’t own the cloud like Amazon does. That’s a big inhibitor.
And now Amazon’s doubling down. It’s going after The Trade Desk with its own DSP [demand-side platform] and Amazon just announced its own supply-side platform (SSP), which will compete with [companies like] Criteo and Citrus.
RTP: This is the new Amazon Retail Ad Service, right? I didn’t initially think of it as an SSP, but that’s exactly what it is.
Burdick: Yeah, Amazon’s providing the [ad] inventory and the whole infrastructure. Now when you create an ad in Amazon, you can click a box that says, “Show up on Amazon’s ad networks.”
This goes back to that head and tail concept of retail media networks. The tail, the ones who really don’t have a right to win, are just launching a retail media network expecting that if you build it, they will come. They won’t come, we know that. But if you launch an ad on Amazon’s retail media network and you then have this ability to say, “Just click a button and also launch on everybody else’s network,” it’s a good value prop for the [retail media networks on the] tail [end].
Another theme at CES was consolidation, because what needs to happen with 253 networks is consolidation. And the whole point of Amazon launching this service is to consolidate the tail.
RTP: What about in-store retail media? There’s been so much buzz around that the last few years, but recently there’s been some more negative press — are companies still excited about that opportunity?
Burdick: Definitely. I mean stores still [represent] something like 80% of sales. For retailers, the store is still king, and that’s definitely part of their retail media value prop. But the focus does seem to be more on owning it instead of using a third party like Cooler Screens. Walmart buying Visio is an example of this. In-store is one part of that unique value prop, full-funnel approach.
RTP: What’s your biggest tip for brands and RMN operators for the year ahead?
Burdick: I think it’s focus; not getting distracted by the shiny penny of the latest AI tool or what your competitors are doing. It goes back to that idea of looking at what you’re doing manually today that you could be automating, and what tools you need to do that. Staying focused on business priorities. It also goes back to the 253 media networks. Do you need them all? No. Focus on your business, your business needs, what are the right retail media networks for you, what are your KPIs and metrics to achieve success?
I also think it’s important to always have a test-and-learn approach because there are so many new things coming out. That’s how you control the “shiny penny” effect — have very specific goals then ask could this new tool help me meet that goal? Yes, okay, I’m going to test it.
Learn more about what executives from the major retail media networks are focusing on in 2025 in this short video, courtesy of Pacvue: