PIVOTREE INDUSTRY REPORT · Q2 2026

When the Shopper Is a Machine

The State of Retail Ecommerce, Q2 2026

A read on what more than 1,600 commerce businesses actually did last quarter, drawn from their hiring, their leadership changes, their mergers, and the platforms they chose to build on.

In a quarter when a new retail AI announcement seemed to land every week, the job retailers posted most often wasn’t an AI job. It was a product data job.

Across more than 1,600 commerce businesses we monitored from April through June, close to one in five opened a role to hire someone whose entire work is product information: the specs, the images, the pricing, the taxonomy, the attributes that sit behind every product a shopper ever sees. That was the largest hiring signal of the quarter.

Everyone talked about the commerce agents arriving at the front of the store. The retailers who understood what was happening spent the quarter staffing the back.

1,600+

commerce businesses monitored, April–June 2026

~22%

posted a product-data role — the #1 signal of the quarter

~12%

carrying active merger or integration work into 2026

Apr→Jun

agentic commerce went from a demo to a checkout button

How we built this

This report draws on Pivotree's retail ecommerce intelligence program, which includes continuous monitoring of hiring activity, executive appointments, mergers and acquisitions, and commerce-platform decisions across more than 1,600 mid-market and enterprise commerce businesses in North America, from April 1 to June 30, 2026. Company-level intelligence is aggregated and anonymized. News related to every individually named company is drawn from public sources — press releases, job postings, and news coverage — and reflects information available as of July 2026.

THE HIRING SIGNAL

Hiring ahead of the machines

Product data isn’t a glamorous hire and usually doesn't come with a press release or a keynote speech. But for four straight months it was the most common role we saw retailers posting, and the job descriptions explain why. These are the people who decide whether a product's size chart is right, whether its price reads the same in three places, whether its category is one an algorithm can actually find.

What retailers hired and restructured for, Q2 2026

Share of tracked retailers showing each signal type. Companies can appear in more than one category.

Product data

22%

Commerce platform

13%

M&A / integration

12%

Executive hires

10%

Order management

Marketplace / channel

The pattern holds at the top of the org chart. The executive hires we tracked skewed hard toward data and AI. The Home Depot brought in Franziska Bell, previously Ford's head of data, AI and analytics, as EVP and Chief Technology Officer. Chewy hired a former Amazon commerce-and-supply-chain vice president, Yunyan Wang, as its CTO. The title on the door increasingly reads data or AI, where a decade ago it would have read IT.

Then there are the roles nobody can fill. Some of the most telling signals in today's retail ecommerce landscape aren't hires at all — they're the same specialized postings reappearing month after month, still open. A senior order-and-integration role we first logged at a major retailer in early March was still live months later, and it's far from unique. Positions like it — the people who make sure an order placed in one system actually reaches the one that fulfills it — now take 90 days to fill if the hiring committee is lucky. The most specialized roles (ones that sit across multiple systems and/or business units) can stay open for six to nine months, according to this year's State of the CIO survey. The work has outrun the talent market. When a company can't hire the person who connects its systems, the need doesn't go away. It just goes unmet, which is a signal all its own.

The Quarter Agents Arrived

Agents don’t shop. They read.

For twenty years, retailers optimized their product pages for a human being who skims and forgives. Someone who glances at a photo, reads half a title, excuses a missing spec because the picture looks right, and buys anyway. Every catalog in retail is built (whether anyone admits it or not) on the generosity of a distracted human.

The machine is not distracted, and it is not generous. When a shopping agent evaluates a product, it parses. It reads the structured data or it reads nothing at all. Every gap a human shopper used to forgive becomes a purchase the agent can’t complete.

When the buyer is an agent, your catalog isn’t describing the product anymore. It is the product.

This stopped being hypothetical over the past year. OpenAI’s Instant Checkout put a buy button inside ChatGPT, with brands like SKIMS among the first you could purchase without leaving the conversation. Google introduced a shared checkout standard — the Universal Commerce Protocol — with Target, Walmart and others, so an agent could complete a purchase from inside search. Target wired itself into both that standard and OpenAI’s. Albertsons put an OpenAI-powered shopping assistant in front of customers across its grocery banners. Levi’s moved its backend onto Azure and began rolling out an AI stylist. The storefront — the thing retail has poured two decades of design into — is becoming one of several ways a product gets bought, and for a growing slice of purchases, no longer the main one.

When the buyer is an agent, the unglamorous hire from a moment ago — the person reconciling attributes and prices — turns out to be the person deciding whether your products exist at all in the channel that's growing fastest. Catalog quality became the price of entry.

Then there are the roles nobody can fill. Some of the most telling signals in today's retail ecommerce landscape aren't hires at all; they're the same specialized postings reappearing month after month, still open. A senior order-and-integration role we first logged at a major retailer in early March was still live months later, and it's far from unique. Positions like it — the people who make sure an order placed in one system actually reaches the one that fulfills it — now take 90 days to fill if the hiring committee is lucky. The most specialized roles (ones that sit across multiple systems and/or business units) can stay open for six to nine months, according to this year's State of the CIO survey. The work has outrun the talent market. When a company can't hire the person who connects its systems, the need doesn't go away. It just goes unmet, which is a signal all its own.

The Consolidation Tax

The deals closed last year. The bill arrives this year.

Retail spent 2025 consolidating. DICK'S bought Foot Locker. 3G Capital took Skechers private in a $9.4 billion deal. Ferrero acquired WK Kellogg for $3.1 billion. Gildan absorbed HanesBrands in a transaction worth roughly $4.4 billion including debt. And in early 2026, Saks Global moved through Chapter 11 and came out rebranded as Exemplar Luxury Group, its debt cut by about three quarters, still holding Saks Fifth Avenue, Neiman Marcus and Bergdorf Goodman under one roof.

The deals were signed in boardrooms but the bill is coming due in the data layer. Every corporate merger is also a systems merger: two companies means two product catalogs, two order-management setups, two sets of integrations built by two teams who never met. Someone has to combine them.

That work appears months after the acquisition press release and begins with the job postings, when the acquirer starts hiring the people who'll spend the next year reconciling part numbers and untangling price books.

Also This Quarter

Three more things worth your attention

Replatforming never slowed down

Roughly one in eight tracked retailers showed a platform-migration signal in Q2. Levi's rebuilt its backend on Azure; others moved between commerce platforms without fanfare. What changed isn’t the volume — it’s the justification. Replatforms increasingly get approved on the strength of data readiness rather than a feature checklist, because the platform is no longer the hard part. Getting clean product data into it is.

Grocery woke up

Grocery has the hardest product data in retail: tens of thousands of SKUs, prices that move constantly, freshness and substitution rules a general-merchandise catalog never has to model. It’s also historically the most cautious corner of the industry about digital. This quarter it moved: Albertsons put an AI shopping assistant in front of customers, Lidl US changed its chief executive, and across grocery and food distribution we saw an unusual cluster of activity in a single quarter. The segment with the most to lose from getting agentic commerce wrong decided it had more to lose by waiting.

The supply-chain seat is turning over

New heads of supply chain and digital showed up repeatedly in our tracking. Leadership turnover in these roles has been a reliable leading indicator that budget is about to move. When a company changes who owns the flow of product and data, it’s usually about to change how that flow works — and the first thing a new leader tends to find is that the underlying data was never built for what they've been hired to do.

	What would a merger or a new channel do to your integrations?
	Most retailers can’t answer this quickly, which is itself the answer. The integration layer is where consolidation and channel expansion turn into either a quarter of clean work or a year of firefighting. The difference is decided long before the deal, in how well the connections are understood.

	Could an agent complete a purchase against your catalog today?
	Not your website — your data. Take your ten best-selling products and check whether their attributes, prices and availability are correct and consistent everywhere a machine might read them. The answer is usually humbling. Better to be humbled by an internal audit than by a channel that simply routes around you.

	Who owns product data as a job, not a project?
	Does anybody own it, continuously, the way someone owns the P&L? On most org charts the honest answer is 'no'. That was survivable when the only thing reading your catalog was a person who'd give you the benefit of the doubt. It isn't a person anymore.

About Pivotree

Pivotree builds, connects and manages the digital supply chain for commerce businesses. We do the work that gets a product from a database to a doorstep. Product data, platforms, integrations and operations, run as a continuous discipline rather than a one-time project. If this report resonated, that's the work we do.

Get a real read on your Agentic Readiness

30 minutes with our retail industry expert Dan Ornstein for eligible retailers, drawing on our work across the industry with brands like Aldo, Princess Auto, Michael Kors, and more.

Pivotree is platform-agnostic, so the conversation will reflect your systems, not a vendor's roadmap.

You’ll leave with a full written Agentic Readiness Assessment and a straight look at whether your systems can actually support agentic commerce — where your product data, integrations, and order flow are ready for agents to act on them, and where they'd break.

Book your strategy call now!

Built for retailers with $250M+ in annual revenue and 10k+ active SKUs — limited spots available.

About Dan

Dan Ornstein helps retailers build loyalty and grow revenue through unified commerce, awesome customer experience, product data, and pragmatic AI. He brings a wealth of experience to the work he does, developed over many years of working with designers, marketers, retail ops, e-commerce and technologists bringing great ideas to life in both the digital and physical worlds.