Retail Pulse Report: AI Recommendations are More Trustworthy Than Store Associates?
Plus, 65% of Americans think we're in a recession (while spending like we're not), while Gen Z adds a new holiday: Summerween
Source: Adobe Stock. It really came through this week, AI-generated against the prompt of merely “Summerween”.
Here’s what to look for this week:
The first signs of tariff-driven inflation
Don’t rule out the possibility of demand-side inflation, if retailers pause orders too hard in the face of uncertainty
Consumers’ reasonable (and crazy) claims about what’s happening in the world (a mid-year take)
ChatGPT and Shopify not really in a battle for checkout supremacy
Summerween, the (new) holiday the Gen Z’s are talking about
Let’s jump in!
Retail Economic Indicators
The biggest news of the last week was the US inflation report, which showed year over year inflation in June hitting 2.7%, an 0.3% increase month of month as well. Fuel prices have been falling, food prices have been increasing, but it was the inflation indicator that excluded food and fuel that was the most interesting, coming in at 2.9% YoY.
CNBC had a good breakdown of why this is most likely attributable to tariff-caused inflation, mostly because the price increases hit a whole bunch of different categories, instead of just very specific ones. As you’ll see below, the UK has been experiencing persistent food inflation while everything else has settled down. In the US, it was a much broader story.
It’s fair to say that this story is not over yet, either. While June numbers started to show some strain, both anecdotal and survey results from businesses show that most retailers and brands have been doing everything in their power in the short term to avoid passing tariff costs onto consumers. Some of that has simply been a matter of trying to contend with wildly fluctuating policy changes that have yielded one shipment with a big tariff hit and another of the same type of good with “normal” tariff costs. But retailers’ ability to continue this practice will come to an end sooner or later.
At a minimum there is a real threat of supply chain contortions causing either over- or under-supply, depending on when consumers go shopping. US apparel imports from China, one of the biggest categories exposed to tariff risks, fell to a 22-year low in May 2025, which was also down 52% from May 2024. I’m sure June was a rush to bring in a ton of merchandise once everyone realized the reprieve. What that means for back to school, or consumer demand and prices in general, remains to be seen. If we go back to too many dollars chasing too few goods – and consumers seem hellbent on continuing to spend – that will be more immediately inflationary than the impact of tariffs.
Meanwhile in the UK, inflation also jumped to an 18-month high in June, of 3.6%. The increase was blamed on rising foot and transportation costs, coupled by a flattening out of energy prices (which had been driving some of the slowdown in inflation over the last few months). Food price inflation has stayed stubbornly high in the UK (4.5%). I’m not close enough to that sector of the industry to know why, but it’s an interesting contrast to US inflation which was more broad-based.
And that inflation reading is important context for understanding sales results. UK retail sales increased 3.1% YoY in June (versus a 0.2% decline in June 2024). The increase in sales was driven mostly by food, health & beauty, and household appliances. But here’s the rub: food sales were up 4.1% year over year – while food inflation was up 4.5%, suggesting that food sales weren’t up because people were buying more, but because food cost more.
Retail Tech & Research Data
I save off a lot of big PDF’s always with the intent to come back and read them in depth, and then they often rack up until I have a big holiday and plenty of time to read them. However, this week was slow for news overall, so I went “last in first out,” and that happened to be The New Consumer / Coefficient Capital’s Consumer Trends mid-year report (which you can find here). I like their take because they ask interesting questions, and so get interesting answers. This was a big, wide-ranging report based on an on-going survey of 3,000+ US consumers. Here are the highlights that jumped out at me:
46% say prices have increased "a lot" in the last 6 months, which is the same as their December 2024 survey
65% believe the US is currently in an economic recession, and 16% think it is "a bad one"
AI awareness is higher than Doritos, TikTok, or Viagra – AI awareness at 98%, Doritos at 88%
More than 60% of Gen Z says it has used ChatGPT over the past month. And 2/3 of students.
Only 17% of Gen Z and Millennials say they are uncomfortable sharing personal data with AI, vs. 38% of Gen X and older
"OpenAI consumer spending is now basically a Paramount+"
Which is over 2x the amount spent on YouTube Premium and 20% of the total spend on Spotify
With retention that falls closer to Paramount+ than Spotify (i.e., less essential than Spotify)
15% of workers are "very concerned" that AI will steal their job before 2030, but it's 27% among those reporting that they are "extremely familiar with AI"
48% of Gen X and older say they won't use AI tools for shopping vs. 20% of Gen Z and Millennials
But among Gen Z and Millennials, 39% prefer a general-purpose search like ChatGPT over 16% who would use a shopping-specific AI
When shopping for supplements, Gen Z and Millennials trust AI recommendations about the same as a store associate (20% vs 19%) – ouch
However, much more likely to trust a store associate for beauty product recommendations (28%) vs AI (20%)
There was a big section on longevity attitudes, as well as how consumers view pets. Americans in general would be willing to spend no more than $100 per month to increase their own longevity, but some would be willing to go up to $200 per month to increase the life of their pet.
And the one that you should laugh at: 35% of Americans say they would exercise at least 2 hours a day in order to live longer. This is your weekly reminder that just because consumers say something, that does not mean they do it.
AI & Retail
Two AI & Retail stories that appear related on the surface but actually not so much. First, OpenAI adds shopping to ChatGPT, through buttons for product searches. When selected, recommended items will direct to the merchant’s website to complete the transaction. The recommendations are bades on a combination of the user’s past preferences plus product reviews from the web.
WIRED asked an interesting question: if affiliate partners make money by putting together “best of” articles that encourage people to click through from the article (something WIRED does a lot of), what happens to the affiliate if ChatGPT uses the content to make a recommendation but bypasses the affiliate’s money-making link (and primary reason for writing a roundup or comparison article to begin with)? OpenAI said they were working on it.
In the meantime, Shopify has added a warning in its code for storefronts stating what bots that might come to scrape the site can and can’t do. The warning states that automated scraping, “buy-for-me” agents, and any end-to-end flow that completes payment without human approval first is not permitted.
At first that could seem to be a counter to ChatGPT’s shopping mode, but Shopify said it was really more to discourage people from going around the standard API’s and SDK’s the company already offers. And when you consider both that a product recommendation could very easily lead to a referral to a Shopify storefront, and also that Shopify has gone in on partnering with OpenAI on embedded shopping from a chat interface, then it becomes far less controversial than it first sounded. They want protection for their merchants against bot depredations, and they want to make sure that Shopify maintains the top of the food chain in what happens after a product is recommended, by pushing third parties to established tools, rather than trying to go around the rules through (still hacky) general-purpose AI interfaces.
Retail Winners and Losers
And last but not least, with a shoutout to one my kids’ most favorite shows, Gravity Falls, Summerween appears to be catching on to the point of being something to retailers are piling in on. The term is the title of a 2012 episode of the show, and celebrates Halloween with summer vibes – think “skeleton beach party” and “jackomelons” and mummy-coated hot dogs (strips of crescent).
A post referencing the “holiday” went viral last year, and this year several products featured by the likes of Walmart and Target have already sold out. With the inventory dynamics of this year, it’s not surprising that retailers would take advantage of the sentiment to presale Halloween inventory.
Halloween is uniquely exposed to Chinese tariffs, with a high dependency on the source for costumes. Plus the date is the date – you miss the inventory window for Halloween sales and there is no making that up, unless you want to pack and hold for next year. And for anyone worried that consumer spending may break by the fall, Summerween offers a chance to cash in on inventory you brought in early, hedging your bets for what may be a slower Halloween (that is also more vulnerable to discretionary spending cutbacks than back to school or holiday).
However, while pumpkin carving tools can easily be repurposed for, say, watermelons, repurposing general Halloween for Summerween would be a lot harder, as there seems to be a demand for pink and blue over orange and green. It is a decidedly more cutesy version of the holiday that most Halloween decorations do not typically embrace. But here’s to the enduring cultural impact of Gravity Falls – perhaps the first Gen Z true nostalgia moment of breathing new life into something that generation genuinely, uniquely grew up with.
If they want to cement the new holiday, though, they’re going to have to agree on a date. Just sayin’.
What Did We Learn This Week?
I do tend to look at both retail sales and inflation together. If retail sales are up, say, 3%, but inflation was up 5% in the same month, I’m not buying that retail sales really grew. Consumers may have spent 3% more than this month the year before but that doesn’t mean they actually bought more – they just paid more. If the gap between inflation and retail sales is too big, then it’s very difficult to claim any retail sales growth at all – inflation ate it all up.
The post-pandemic years saw a lot of months where that kind of gap occurred. Things then seemed to settle down, but June suggests we’re back to watching that gap, and in the US, the main cause of that gap is tariffs. However, I think it’s very important to keep in mind two things: one, as economists have been warning, tariffs do not directly, immediately lead to inflation. And consumers for certain have discounted that tariffs’ costs are going to bite – which would definitely be a miscalculation. It’s too early to tell.
But two, tariffs did cause retailers to put the brakes on imports, and in some cases, very hard. It’s not clear that they were able to make that up to bring in halted orders during the (first) reprieve or not. And too many dollars chasing too few goods is directly inflationary, as we’ve learned over the past four years.
Also, consumers are both still grumpy and vibe-cessionary – believing that things are way worse off than they actually are, and yet still spending like drunken sailors on payday, as splurges on a made-up holiday that hearkens back to their tender youth prove out (I’m talking about Summerween here, even if you or I don’t really get it).
All that to say, consumers are complicated. They definitely are not the “rational, self-interested actor” of the economics of old. We’re seeing that play out live right now. But the most important thing to remember is that it may be live, but the data is not real time – it lags, and there is a greater lag between when initial shocks come and the full impact shows up in the data than I think most people credit. And we’re in one of those lags right now.
Until next week!
- Nikki
I write this weekly article as much to force myself to make sense of what I read on a daily basis as to say anything about it publicly. I will continue to do so - for free - for as long as I can sustain it.
If you like what you’ve read and you haven’t subscribed, please do! It helps me know I’m saying interesting things.
If you’re already subscribed but you can think of someone who would enjoy reading this, please share the article.
Or better yet, share the whole publication and let them get in on the weekly fun!
Thanks for reading, and thanks for your feedback and support!