Last week, the news became official: DoorDash acquired Instacart for $17 billion.
Earlier in the week, when the deal was still a well-sourced rumor, Needham, an investment bank, put out a note regarding the deal. Needham’s working theory was that the deal is about combining the two advertising businesses. Everything else is logistics.
So what does this deal actually mean for commerce media?
The easy read is that a delivery app is buying another delivery app. Actually, DoorDash is buying a revenue mix it has never managed to build for itself. Not for lack of trying. DoorDash bought ad-tech company Symbiosys for $175 million in 2025 to take its ads beyond its own app, and it has spent years pushing into grocery and retail.
Buying Instacart is DoorDash’s admission that building on its own wasn’t getting there fast enough.
The margin
Here is the arithmetic. Instacart’s advertising and other revenue was $297 million in the second quarter, 28.5% of everything it earned. Call it more than a quarter of the company. DoorDash passed a $1 billion run rate in ads two years ago and hasn’t broken the number out since. By my count, that’s around 10% of its revenue.
Both companies have billion-dollar ad businesses. Only one of them is an ad business.
That matters because delivery is a hard business and advertising is a highly profitable one. According to Wedbush, a financial securities firm, Instacart brings more than grocery drop-offs; it brings a marketplace across many grocers, plus an ad business brands pay to rank higher in search.
Delivery is low margin, and ads are high margin. Ads lift what the platform earns per order without charging the customer more. That is the beauty of a blended P&L.
The front door
How did Instacart end up setting the bar? It became an ads business because the grocers couldn’t deliver what they had promised brands.
Retail media was pitched as performance media for commerce. Ads at the point of conversion, closed loop, every dollar accounted for. What brands got from most grocers was run-of-site advertising with a sales report attached.
Some grocers built serious networks, such as Kroger Precision Marketing and Albertsons Media Collective. But even the good ones could only reach one retailer’s shoppers, and that left brands stitching together a dozen buys to cover a category.
Instacart could do something the grocers couldn’t. Because it sat on top of the last mile, it could bundle baskets and get hyper-specific across hundreds of retailers in one buy. A brand didn’t have to negotiate with Kroger, then Costco, then Publix. It bought the category once.
That’s the Uber model – a network of independent operators behind one front door – applied to grocery. (Not so coincidentally, Uber was also reportedly in the mix to acquire Instacart, but it backed out.)
The combination
Together, DoorDash and Instacart reach across occasions no one else covers in one place. DoorDash’s share of US delivery spending has grown from 44% to 59% over five years, while Instacart’s fell from 26% to 18%. Combined, a CPG brand could find the same household ordering dinner on DoorDash on Tuesday and groceries on Instacart on Saturday, across restaurants, grocery and convenience in a single buy.
But Instacart has differentiators that could prove to be worth its $17 billion price tag. Instacart is not reliant on individual grocery retailers, because it has direct relationships with consumers. Consumers opt in to sharing loyalty card data with Instacart for rewards at the grocer, and Instacart carries that opt-in. Plus, Instacart operates its own planogram model, so it decides how prominently it features products based on its own brand deals.
All of those factors make Instacart a true commerce media business. And Instacart’s direct relationships with consumers and brands can’t be easily disrupted by a grocer deciding to launch its own retail media platform.
Meanwhile, DoorDash built its relationships directly with restaurants. So Instacart’s relationships with grocers are truly additive.
The proof
Now, here’s the uncomfortable part: attribution.
Instacart nailed attribution on its own app. It has a 14-day window, basket-level reporting, a clean room since January and MRC accreditation for its impressions and clicks. What it never nailed is attribution to the retailer’s total sales, which is what the grocers promised in the first place.
Instacart sees the Instacart basket. Most grocery still happens in a store, and that halo is modeled, not measured. Brands kept spending anyway. They pay for the basket, not the proof. What is interesting is that the burden of proof shifted to the retailer rather than Instacart. Instacart can prove what happens in its own basket. So when a brand asks what its spend did in the store, the retailer is the one left without an answer.
There’s a bigger point here for commerce media. We’ve spent five years calling it an emerging channel, the thing on the test budget. It has just been priced as a scaled business asset, with closed-loop measurement on its own basket and modeling to show brands its impact in-store.
A public company just paid $17 billion, about a 30% premium, largely for an ad business built on shopper data. And the analysts covering the deal are valuing that ad business, not the drivers.
Commerce media is no longer just a line in the marketing plan; it is a line in the acquisition model.
“Data-Driven Thinking” is written by members of the media community and contains fresh ideas on the digital revolution in media.
Follow Kristina Kaganer, Auxbow and AdExchanger on LinkedIn.
