10 minutes of expert insights every weekday. Your morning ritual for staying ahead in retail media.
BFCM Bifurcation: Consumers Split On Agentic & In-Person Shopping In 2026
===
[00:00:00] Kiri Masters: A lot of people listening to this podcast spent the 2010s trying to talk shoppers into buying stuff online. I actually launched an e-commerce business in 2012, and I spent [00:00:15] years after that helping brands grow online on marketplaces like Amazon. So I'm really feeling those gray hairs popping through when the generation that has been raised on screens now [00:00:30] treats a Black Friday trip to the mall the way it treats vinyl and paper magazines, with borrowed nostalgia for ancient rituals of their forebears.
[00:00:43] What I'm talking about here [00:00:45] is this changing attitude and behavior around Black Friday, Cyber Monday, and trends that we've seen over the last couple of years moving in a direction that I think a lot of people [00:01:00] wouldn't have expected, which is that while online sales are the highest they've been as a percentage of US retail We're also seeing shoppers change their buying habits around [00:01:15] this major spending holiday really significantly as well.
[00:01:19] So the online sales now is up to 17.1% of US retail in the second quarter of 2026, up from [00:01:30] 16.3% the year before
[00:01:31] What I think we're going to see next week over the Black Friday, Cyber Monday shopping period is two seemingly disparate types of behavior. One, [00:01:45] shoppers calling in their AI sidekicks before and after they buy, while also many people also planning to do a lot of their shopping in person, and perhaps even all of these [00:02:00] things at once for different types of missions and experiences throughout the shopping holiday.
[00:02:07] Today, I'm going to round up three or four different pieces of research and announcements that popped on my desk [00:02:15] over the last few days that I think is really pertinent going into next week
[00:02:19] So let's jump in
[00:02:20]
[00:02:22] Kiri Masters: the first trend here is agents everywhere except for the register. So while a lot of the [00:02:30] industry's agentic shopping work and commentary is aimed at the checkout, there was a November survey from the company Riskified, which found that US shoppers are nearly twice as likely to have used AI to [00:02:45] handle a return or a refund as they were to have completed a purchase directly through AI.
[00:02:53] and Nearly half of survey respondents expect the AI platform, not the retailer, to [00:03:00] issue the refund if an agent buys the wrong thing.
[00:03:04] And as we know, purchases made over the next couple of weeks have return behavior that lasts well into January, where [00:03:15] people are buying gifts and they get returned well after the Christmas period. So that expectation and how, you know, these returns play out at scale is, is going to kind of be a [00:03:30] first for, for this year and into 2027.
[00:03:33] What's interesting from a retail media standpoint is that-
[00:03:37] A lot of the performance that we'll see in retail media over the next few weeks won't account [00:03:45] for those returns in January
[00:03:49] So I think it's gonna be really important, if not difficult right now without the ultimate suite of tools and resources that have been developed to solve [00:04:00] and track this problem, to figure out how many purchases were made by agents versus how many purchases were made by humans, and how that return and refund profile, how different does [00:04:15] that look?
[00:04:15] So I'm very interested if there's any sources out there that are tracking that data, I'd love to hear about it. This finding pairs really well with another report that came out recently from [00:04:30] Accenture, which was titled Talk to My AI Agent, and it was part of its twenty twenty-six Consumer Pulse research.
[00:04:37] And there was a fascinating graph in this report which shows... It's a line graph. It [00:04:45] shows the propensity to delegate actions to AI agents across the consumer journey. So it breaks it up into about twelve or so stages of the consumer journey, the path to purchase, the purchase itself, and then [00:05:00] post-purchase.
[00:05:00] So it's actions like prompting the need, discovering options, deciding where to buy, managing customer service, managing loyalty, et cetera. And so Accenture in their survey, [00:05:15] scored how willing consumers are to hand each step of that process to an AI agent. And so what you see in this profile of this line chart is willingness to outsource is the highest at the start [00:05:30] when it is all about working out what you need, and it stays high after the purchase, which is managing that post-purchase experience, like booking the delivery date, And all of that.
[00:05:41] And but it bottoms out, has the least [00:05:45] likelihood of being passed to an AI agent at payment decisions. So shoppers will let an agent find the options and chase the refund, but they wanna be the one who pays, which at this point is completely understandable [00:06:00] while all of those payment systems are being figured out and the, the card companies are getting in there and deciding how they're going to handle things like fraud, and the AI [00:06:15] companies are building wallets.
[00:06:16] There's so much happening in this space, but consumers really haven't caught up to what's available and what they feel comfortable with yet. [00:06:30] Costco isn't scaling its RMN with legacy ad tech and manual solutions. It's leaning into a cloud-centric composable stack. As part of the Costco Velocity [00:06:45] Network, GrowthLoop empowers the retail media team to build audiences directly from the Data Cloud, enabling faster activation, greater relevance, and better performance, all while [00:07:00] maintaining privacy and governance standards.
[00:07:03] The result? Exceptional value and experience for Costco members and better ROI for brand advertisers. Learn why the [00:07:15] GrowthLoop Composable Commerce Media Solution was the right choice for Costco's Velocity Network. Visit go.growthloop.com/breakfast. That's [00:07:30] go.growthloop.com/breakfast
[00:07:32] Making a little bit of a pivot here to talk about Pinterest, and, I've, I heard a lot of skepticism about whether Pinterest has a real role in AI shopping, [00:07:45] but they recently released a suite of features that they've bucketed under, Pinterest Intelligence.
[00:07:54] And to me, it looks quite thoughtful and instructive because [00:08:00] it is v-a very visually native system that's built for the part of gift shopping and holiday shopping that text-based chatbots really, you know, can't deliver a very [00:08:15] inspiring format for discovery. But Pinterest is kind of born for that, right?
[00:08:22] Um, so this is... these are more visual native AI experiences that are suited to the casual browser. [00:08:30] So it's things like, um, deal alerts on your wish lists, and those wish lists obviously span multiple brands and retailers. And so I, I love the concept of this rather than saying [00:08:45] RIP to my inbox when I sign up for a dozen brand email lists to get the alerts about their new drops and discounts and things like that.
[00:08:54] This is actually a really great place to aggregate all of those brands that I follow and products that [00:09:00] I'm interested in, and aggregate that inspiration and be notified of, new products and things like that. So it's, look, it's, it's not a monumental mind-blowing step, but I think it's showing us [00:09:15] what a more visual format of AI-enabled shopping could look like So, um, kind of hard to describe on a podcast, but check out the accompanying blog today to see some images of that.
[00:09:28] And then finally, where [00:09:30] retail media fits into all of this. The ad intelligence firm Guideline tracks, ad spend through agency billing data. And so that's a really great data source that I've [00:09:45] recently, come across. And so it's really tracking, like, the larger ad spenders, uh, through the, the big agency billing cycles.
[00:09:53] And so Guideline is projecting that November and December ad spend will reach a [00:10:00] record this year, a record of $25.11 billion, up from, about a billion dollars, uh, higher than last year. So that's rec- that's a record in dollars spent, but it is a slowing [00:10:15] growth rate. And so I think this is useful to know.
[00:10:18] Advertisers are, you know, bullish
[00:10:21] So where does this all leave us in the ad world? Well, I think the default holiday plan sends a big [00:10:30] share of ad budget to things like sponsored products and ads that sit right there before the transaction. But that ad depends on the shopper actually being on the retailer's site at the moment of purchase.
[00:10:43] And this year, a lot more [00:10:45] shoppers are planning to be standing in an aisle ~and using third-party chat assistance~
[00:10:49] ~No, I'll just... Sorry, I'll just leave that ~the National Retail Federation's holiday survey found that fifty-three percent of shoppers plan to buy online this year. So that's, you know, n-n-north of half. [00:11:00] But, and so online is still the most popular destination, but that is actually the lowest share since twenty fifteen.
[00:11:09] I've argued before that in-store retail media is one part of the business that [00:11:15] AI can't bypass at all, and a store-heavy Black Friday also puts the rest of retail media's job back in focus, that we should be building awareness in the [00:11:30] weeks before and win- winning consideration at the shelf on the day.
[00:11:35] There was a holiday campaign from Sephora in collaboration with Pinterest that demonstrates how those two objectives can [00:11:45] play together quite nicely. So on Pinterest, Sephora is running what they're calling Premier Spotlight video takeovers that guarantee the first impression when you open the [00:12:00] app. And offline, it's the exclusive beauty partner in some co-branded digital out-of-home placements across Westfield malls in New York, LA, Chicago, Miami, [00:12:15] and Atlanta So really seeing Sephora play into this dual role as a online and offline destination leading up to the event.
[00:12:26] So next week, next week I think we'll be spotting [00:12:30] some entirely new shopper behavior that we haven't seen before, or at least has been very nascent. We're going to see shoppers asking agents to build their wishlist, find the best deals, but [00:12:45] then also driving to the mall to pick it up in person on Black Friday.
[00:12:50] And then they'll ask the agent to handle the return for them in January. So let's see how this all plays out. [00:13:00] What a time to be alive in e-commerce and media. I do have some more projections for Black Friday, Cyber Monday coming out over at my column for The Drum later this week with some projections from Salesforce and some other [00:13:15] analysis.
[00:13:15] So if you're following me on LinkedIn, you should be able to see that. And finally, just wanna call out, there is a new episode of the Adgentic podcast, which is my new podcast on the Refi by Media Network. I'm interviewing [00:13:30] Deborah Aho Williamson, who writes the newsletter The AI Ad Economy,
[00:13:35] And I loved talking to Debbie about this because she was there at the beginning of social media advertising. She was an analyst at eMarketer at that time, [00:13:45] and a lot of the things that she's seeing and hearing about are very familiar to the dawn of social media advertising. So we get into all of that and look at how history seems to be repeating itself in some ways with [00:14:00] ads in AI ecosystems.
[00:14:02] So make sure that you check out that podcast wherever you get your podcasts that you're listening to right now. We're also on YouTube and on LinkedIn newsletters. We'll link up to all of that in the show notes. [00:14:15] Thank you for listening, and I'll catch you tomorrow
[00:14:17]