Key Takeaways
- On Wayfair's Q1 2026 earnings call, CEO Niraj Shah committed to engaging early with every major agentic AI platform and helping shape its direction
- The retailer is partnering with Perplexity, OpenAI and Google in parallel, with Google Gemini already running checkout through the Universal Commerce Protocol (UCP)
- The strategy splits cleanly into an advertising track (Meta, Pinterest, Google) and a purchase-experience track (ChatGPT, Gemini), backed by first-party AI work inside the company
What "Be Everywhere" Actually Means
Its CEO said Wayfair is working with Perplexity, OpenAI and Google on agentic commerce in addition to other AI offerings.
www.digitalcommerce360.comOn the Q1 2026 earnings call reported by Digital Commerce 360 on May 5, 2026, co-founder and CEO Niraj Shah laid out Wayfair's posture toward agentic commerce in a single sentence. "We want to be everywhere. We want to be there early, and we want to help shape the direction. That's the way we think about agentic commerce." That line captures the future the leading North American home furnishings retailer is now planning around.
Wayfair sits at No. 11 in the Top 2000 Database of North American online retailers and ranks first among Housewares & Home Furnishings sellers. When a player of that size declares it intends to engage with every major AI platform and co-shape the standards, the move carries weight beyond its own catalog.
What is striking is that Shah's framing avoids dependence on any single platform and deliberately separates advertising from purchase experience. He said Wayfair has been an early partner with Meta, Google and Pinterest in developing ad units, with those efforts still in beta. The purchase side, by contrast, runs on dual rails: integrations with external surfaces such as ChatGPT, plus first-party AI work inside Wayfair's own properties.
Three Partnerships, Three Different Roles
Each of the three AI platforms named by the CEO plays a distinct role.
The most concrete is Google. At NRF on January 11, 2026, Sundar Pichai announced the Universal Commerce Protocol (UCP), and Wayfair was a co-developer alongside Shopify, Etsy, Target and Walmart. UCP is an open-source standard, primarily driven by Google and Shopify, that defines how AI agents and ecommerce platforms interact. It covers the four operational stages of agentic commerce: discovery, capability negotiation, checkout and post-purchase handoff.
By February 2026, UCP-powered checkout was live in Google's AI Mode and the Gemini app, and Wayfair products could be purchased without leaving Google. Shopping inside Gemini turned, for Wayfair, into a transaction completed entirely inside an AI search result.
The OpenAI partnership uses ChatGPT as a purchase touchpoint. CFO and CAO Kate Gulliver described "off-site shopping through recent retail integrations with platforms such as ChatGPT," language the industry reads as discovery integration via OpenAI's Agentic Commerce Protocol (ACP). Co-developed with Stripe, ACP uses Shared Payment Tokens (SPT) to complete in-chat transactions without exposing payment credentials.
The Perplexity partnership is the earliest stage of the three but has a distinct character. Perplexity's strength is the research-to-recommendation phase of AI search, and it has offered "Buy with Pro" through a PayPal partnership since November 2024. For Wayfair, that means securing presence at the moment users are comparing interior options, well before the purchase decision is made.
A Two-Front Campaign: Ads and Purchase
Read carefully, Shah's comments describe a deliberate two-front campaign.
On the advertising front, Wayfair is co-developing AI-era ad units with Meta, Google and Pinterest. The aim is to secure ad surfaces in the moment an AI agent interprets user intent, so Wayfair products surface as candidates. The Pinterest work parallels its shoppable TV initiative, and the Google work extends in the same direction as Ulta Beauty's Gemini agentic commerce launch.
On the purchase front, ChatGPT and Gemini themselves are treated as customer touchpoints. The investment is in keeping the catalog verifiable, discoverable and transactable on every AI surface. Whichever AI a user chooses, Wayfair products should appear accurately as options and the purchase should complete in place.
A third layer, often overlooked, is first-party AI inside Wayfair. Gulliver said the company uses AI to improve merchandising, noted that engineers have used "various forms of machine learning for years," and explained that AI accelerates how customers discover and engage on the site. External exposure and internal experience must move together; either alone is insufficient.
Where Agents Will Win, and Where They Will Not
A revealing moment on the call was Shah's clear separation of categories where agents will and will not change behavior.
He named three categories where agents will matter: replenishable items (paper towels, dish soap), commodities (phone chargers, where price and quality matter and brand does not), and technical items (premium televisions, where specs are comparable and consumers are brand-agnostic). All three lend themselves to AI judgment along price, specification and delivery axes.
Fashion, beauty and home are different. Shoppers learn through the act of shopping, emotion plays a heavy role, and people actively do not want to own the same things their neighbors own. That cultural reality, Shah suggested, limits how far agents can substitute for human discovery in those categories.
He did acknowledge that low-price furniture, like a barstool, will face agent-mediated competition with Amazon, Walmart, Target, Temu and TikTok Shop. But he dismissed that volume bluntly: there is no margin in it, and it is not where retailers differentiate. Wayfair's emphasis stays on the higher-consideration end of home.
Four Lessons for Ecommerce Operators
Wayfair's playbook offers structured lessons for retailers operating at very different scales.
The first lesson is that catalog readiness for multiple platforms is the foundation. Wayfair feeds the same product data into Google's UCP, OpenAI's ACP and Perplexity's surface. Underneath sits a single source of truth for product attributes, inventory, pricing and shipping, normalized into each protocol on the way out. Wayfair already runs agentic AI for catalog enrichment in the U.K., correcting product attributes across tens of thousands of SKUs.
The second lesson is to evaluate ads and purchase separately. Agent-era ad units are largely in beta, and measurement models differ from traditional CPC. Shah called the Meta, Pinterest and Google work an early-partner beta. Treating it as a learning investment, while expecting purchase-experience integrations to carry near-term economic returns, keeps capital allocation honest.
The third lesson is category-aware prioritization. The replenishable, commodity and technical segments Shah identified are where agentic commerce arrives first. Operators whose flagship categories sit there should be moving on UCP and ACP support now. Operators in fashion, beauty or higher-end home need a different muscle: how to express comparison-resistant emotional value through brand storytelling, styling and curation, on AI surfaces that are tuned to compare specs.
The fourth lesson concerns loyalty and Merchant of Record. Shah was direct about suppliers going around retailers: in home furnishings, where logistics and damage are real, supplier direct-to-consumer rarely works economically. The deeper point is general. As long as retailers can differentiate on customer service and logistics, they retain a role even when discovery moves to agents. He framed the loyalty program explicitly as a way to avoid paying advertising costs repeatedly, pointing toward CRM-led direct relationships as the long-term lever.
Closing Thoughts
Wayfair's "be everywhere" strategy avoids betting on a single AI platform and instead builds parallel relationships with Perplexity, OpenAI and Google so that the retailer has presence on every major surface. UCP-powered checkout is live in Gemini, ACP discovery integration is progressing in ChatGPT, and Perplexity engagement is in place at the early-partner stage. Given that the standards battle is unresolved, this triangulated approach is the most risk-hedged stance available.
In parallel, internal AI investment proceeds: catalog enrichment in the U.K., French localization for Quebec, and faster product onboarding all contribute to operating leverage. Q1 2026 came in at $2.9 billion in revenue, up 7.4 percent year over year, with a five-year-best Q1 adjusted EBITDA margin. Announcing this strategy from a position of recovering fundamentals, before the platform standards settle, is a deliberate choice to be on the shaping side rather than the receiving side.
The takeaway for ecommerce operators is straightforward. Agentic commerce traffic, as Shah noted, is still very small in absolute terms. But the operators who invest now in structured data, multi-protocol readiness and category-by-category prioritization will be the ones AI agents select once that traffic scales.
