Hello GrocerTalk listeners,
Well, we’re back.
After kicking off GrocerTalk last week with Chris Walton, I’m delighted to welcome another retail veteran to the show.
This week, I’m joined by Martin Bailie, a lifelong grocer who has held senior leadership roles across three continents, including CEO roles at Tesco and Tata in India, Group COO at Primark, and CEO of Lidl UK & Ireland.
So, yes, there was plenty to talk about.
And this week’s episode gave us a particularly interesting mix of stories.
Target may finally be showing signs of a turnaround. Sainsbury’s has paused its facial recognition rollout after a shopper was wrongly identified. Amazon and Uber are both betting heavily on drone delivery. And Ocado is quietly changing the way it sells its technology.
Underneath all of those stories, though, I think there’s a bigger theme.
Retailers are increasingly trying to build flexible operating models that combine technology, people, physical stores and digital channels rather than relying on one answer to solve everything.
Here’s what caught our attention.
Is Target Finally Turning a Corner?
We started with the story that really caught my eye this week: Target’s Q2 results. Net sales were up 5.3%, comparable sales increased 3.8%, traffic was up 3.6%, digital sales grew 8.7%, and same-day delivery grew more than 25%. Even more interestingly, grocery sales were up 7.2%, compared with less than 1% growth in the same period last year. Target has also raised its full-year sales guidance to around 5%.
There’s just one problem. Target has had a pretty brutal couple of years, so it’s fair to ask how much of this is genuine momentum and how much is simply a retailer recovering from a weak base. Martin Bailie was positive, but appropriately cautious. His view was that the headline deserves the headline. Target is now growing alongside Walmart and Costco, two retailers that you wouldn’t necessarily have put it alongside a few years ago. But he also pointed out that Target benefited from almost $1 billion in tariff refunds, meaning the profitability story is slightly less impressive once that windfall is removed.
For me, the more interesting part of the results is what’s happening underneath the headline. I think there are three things to watch: store standards and operations, merchandising, and grocery. Traffic is up, suggesting customers are noticing improvements in the physical experience, while Target needs to bring back some of the excitement and discovery that historically made the brand distinctive. And food and beverage is growing rapidly, which matters because Target increasingly needs grocery to become a bigger part of the reason people visit its stores.
The next big test is back to school. If Target can continue to improve its grocery proposition while getting fashion and general merchandise moving again, then this starts to look much more like a genuine turnaround. One quarter is not enough, but it’s certainly more interesting than it was a year ago.
Sainsbury’s, Facial Recognition and the Human Problem
Our second headline took us in a very different direction. Sainsbury’s has paused the use of AI-powered facial recognition in one store after a shopper was wrongly identified as someone previously associated with theft or violence. The technology itself is reported to have an identification accuracy rate of 99.98%, which sounds pretty impressive until you think about how many people walk through a supermarket every week.
And this is where the story gets complicated. Martin Bailie’s view was that the technology itself isn’t necessarily the problem. The bigger issue is what happens when technology hands a decision to a human being. In this case, Sainsbury’s said the mistake was caused by human error rather than the facial recognition system. Martin’s point was simple: if you introduce new technology but don’t also change the processes, training and systems around it, you haven’t actually transformed the operation. You’ve just added another piece of technology.
There’s also a much bigger reason why this matters. Retail workers are dealing with extraordinary levels of abuse, violence and theft, and Martin pointed to the growing pressure on store colleagues and the responsibility retailers have to protect both employees and customers. So the question isn’t simply whether facial recognition is good or bad. It’s whether retailers can use technology to protect people without creating a new set of problems for the very people they are trying to protect.
As Martin put it, the goal should be about “protecting people, not policing them.” I think that’s an important distinction. Retailers are going to identify customers in stores in more ways, whether that’s for security, loyalty, personalization or retail media. The technology is not going away. The challenge is making sure the human layer keeps up with it.
Amazon and Uber Are Betting on Drones
We then broke our own rules slightly. For headlines three and four, we decided to talk about two drone stories together. Consider it a GrocerTalk buy one, get one free. First up, Amazon is expanding its Prime Air drone delivery service to nearly 500 cities and towns by the end of 2026, representing a sixfold increase from its current footprint. Amazon says its drones can deliver items weighing up to five pounds, with more than 60% of its most frequently purchased products falling within that range.
The really interesting thing, though, isn’t the five-pound limit. It’s the scale of the ambition. Martin Bailie described the announcement as a shift from drones being a novelty to becoming part of the delivery infrastructure. Walmart has already been pushing hard into drone delivery, and now Amazon is clearly responding. Both businesses are telling us the same thing: speed matters. Walmart recently reported that more than a third of its store-fulfilled e-commerce orders are now delivered within three hours, and drones become another piece of that puzzle.
But Amazon wasn’t the only company making a big drone announcement this week. Uber has announced a strategic partnership with Zipline, including an investment in the drone delivery company, with the ambition of reaching one million drone deliveries per day by the end of 2029. That is a very big number, and Martin was particularly interested in the difference between the two strategies. Amazon is effectively building the airline. Uber is buying a ticket on someone else’s plane.
Amazon is investing in the infrastructure, technology and fleet. Uber is doing what Uber does best: building a marketplace around somebody else’s physical infrastructure. I think that distinction is fascinating. Uber doesn’t need to become a drone manufacturer. It just needs to make drones another part of its platform. And with DoorDash also moving into the space, there is clearly a race developing. The obvious question is whether we actually want millions of drones flying around our cities. Regulation will almost certainly become a much bigger part of this conversation, but for now, the message from Amazon, Walmart, Uber and others is pretty clear: drone delivery is moving from experiment to infrastructure.
Ocado Is Selling a Menu, Not a Model
Our fifth headline took us back to e-commerce fulfilment. Ocado has confirmed that its latest European technology customer is Danish online grocer Nemlig. Ocado will build an automated customer fulfilment centre near Copenhagen, including its latest technology, while Nemlig will continue to operate its existing online shop and last-mile delivery operation. On the surface, it’s a fairly straightforward customer announcement.
But Martin Bailie saw something much more interesting underneath it. Ocado is no longer simply selling one giant automated fulfilment model. It’s increasingly selling a menu. And I think that’s an important distinction. The idea that every retailer should choose one fulfilment model and stick with it has always been questionable. The right answer might be store picking in one catchment, a micro-fulfilment centre somewhere else, and a highly automated customer fulfilment centre where order density and economics make it worthwhile.
We’ve talked about this before on GrocerTalk, and the Kroger example is particularly interesting. Kroger hasn’t simply abandoned Ocado. It has adjusted its approach while continuing to use the technology in parts of its network. At the same time, it is expanding its use of Instacart-powered store picking, including combining grocery and pharmacy orders into a single delivery. That’s the bigger story. There isn’t a single answer to online grocery fulfilment.
The winners will be the retailers that understand their individual economics, catchments and customer needs well enough to build the right mix. Store pick, MFCs, CFCs, automation, partners and technology. It’s all about the blend. And I think that’s increasingly true across retail, not just fulfilment. The best operators aren’t necessarily choosing one technology or one model. They’re working out how all of these pieces can work together.
Then We Got to Grab & Go
Before letting Martin Bailie go, we finished with GrocerTalk Grab & Go. Martin brought a story about Gopuff launching its first seasonal private-label range under its Crave brand, including cakes, cookies, danishes and cookie dough. It might sound like a relatively small announcement, but I think there’s something interesting happening here. As quick-commerce businesses mature, they’re moving beyond simply being delivery platforms and starting to think much more seriously about merchandising, assortment and private label.
Amazon has been doing this for years, and now Gopuff is moving in the same direction. If quick commerce is going to become a genuine grocery proposition rather than simply a convenience service, private label will surely become an increasingly important part of that equation. Martin’s point was that private label is a trend that isn’t going away, and I think that’s particularly interesting as newer grocery models start to mature.
My Grab & Go story was slightly more nostalgic. It’s scarcely believable that I’m saying this in 2026, but Walmart and Sam’s Club are finally starting to accept Apple Pay. Yes, really. Walmart has resisted NFC payments for years, even developing its own alternative, but that finally looks set to change, with stores expected to support NFC payments by the end of this year.
Martin then brought us back down to earth with a useful counterpoint. Instacart’s Q2 orders were up 9% and revenue was up 14%, but net income was down 4%. Growth is great, but growth at what cost? Which, in many ways, brought us right back to the Target conversation we started with. Revenue growth matters, but operational efficiency matters too. And increasingly, retailers are going to have to deliver both.
Want more retail insights like this? Subscribe to GrocerTalk to stay ahead of the headlines every week wherever you get your podcasts:
Apple Podcasts | Spotify | SoundCloud | Amazon Music
Be careful out there,
Ben and the entire Omni Talk Team
Music by hooksounds.com.
Omni Talk® is the retail blog for retailers, written by retailers. Chris Walton founded Omni Talk® in 2017 and have quickly turned it into one of the fastest growing blogs in retail.