There’s a lot of noise right now about AI and product listings. Most of it is either “AI will write your content for you!” (it won’t, not the parts that matter) or “the sky is falling, search is dead” (it isn’t, but it’s changing). Between those two extremes is the actual work.
For brands selling across major retailers, the shift from keyword-based search to AI-assisted and generative search is happening faster than most teams are adapting to it. Here’s what we’re seeing work across Amazon, Walmart, and Home Depot right now - and what we’re telling our clients to prioritize.
Amazon Rufus is the most immediate pressure point
Amazon Rufus - the generative AI shopping assistant - reached roughly 300 million active customers through 2025 and, per Amazon’s Q4 2025 earnings call, is already driving an estimated $12 billion in incremental annualized sales. More importantly for brands: Rufus users are converting at a ~60% higher rate than non-Rufus users.
That changes the math on listing optimization. Rufus doesn’t just match keywords against titles. It reads the whole listing - title, bullets, description, A+ content, reviews, and Q&A - and makes a recommendation based on what the customer is actually asking.
What that means in practice:
- Listings written for humans are winning. Keyword-stuffed copy that passed the A9/A10 era performs worse in Rufus. Clear, use-case-driven language wins.
- Q&A is the single most undervalued real estate on the PDP. Rufus pulls from Q&A more than any other listing element. If your top 20 SKUs don’t have active, on-message Q&A, that’s a two-week project with outsized return.
- Use-case specificity beats spec-dump titles. “32oz stainless tumbler for daily commuter” outperforms “32oz stainless tumbler - leakproof, insulated, dishwasher safe, BPA free.”
- A+ content is no longer optional. It’s now a core input into the model that decides whether to recommend you.
Sellers who get in front of this in 2026 will hold a 12–18 month visibility advantage over competitors who haven’t adjusted.
Walmart is moving in parallel - and for merchants, faster
Walmart has gone further, faster, than most brands realize. The customer-facing generative search is now live across iOS, Android, and walmart.com, and it’s built to understand query context, not just keywords.
On the seller side, Walmart Connect’s Success Hub now offers AI-powered content suggestions tuned to Walmart’s own content standards - essentially, a reviewer’s checklist operating in real time. Their merchant-facing assistant, Wally, is being used internally for item setup, data entry, and calculation-heavy merchandising work.
What to do about it:
- Treat Walmart content as its own product. Copy that works on Amazon often fails Walmart’s content quality scoring. Different rules, different structure, different judgments on what’s “promotional.”
- Use Walmart’s AI suggestions, but verify. The tool is useful. It’s not a substitute for a human who understands the category or the buyer.
- Prioritize listing completeness. Walmart’s quality score has always rewarded complete listings. With gen-AI search layered in, incomplete attributes don’t just hurt SEO - they hurt whether your product gets surfaced at all.
Home Depot is the one most brands aren’t watching closely enough
In January 2026, Home Depot expanded its partnership with Google Cloud to deploy agentic AI across digital, stores, and supply chain. The most important move for brands: AI-generated materials lists on the Pro platform. A contractor describes a project in voice or text, and Home Depot’s AI returns a grouped bill of materials.
If your product isn’t structured to be recognized as part of a project - not just a SKU - you won’t show up in that list. That’s a different optimization problem than PDP keyword rank, and most brands don’t have anyone working on it yet.
What to prioritize:
- Project-context metadata. Attribution, application, compatibility - the fields most brands leave thin or inconsistent.
- Pro-focused content. The Pro segment is where Home Depot’s AI investment is most concentrated. B2B-adjacent content, installation specs, and compatible-product signals matter more than they did in 2024.
- Relationship with the Home Depot content team. AI is being layered on top of existing content standards, not replacing them. The retailers that are farthest along with AI are still evaluating human-created structured content as the foundation.
The honest caveat
AI tools are not a shortcut to good listings. They’re an amplifier of whatever foundation you already have. Clean structured data, thoughtful copy, real Q&A, and complete attribution all compound. Thin, inconsistent, or keyword-stuffed listings will be exposed faster than ever.
What this actually requires is operational discipline across every retailer you’re on - which is precisely the work that gets deprioritized when an internal team is stretched thin.
The takeaway
The brands winning in AI-assisted retail search aren’t the ones running the most AI tools. They’re the ones doing the unglamorous work of fixing listing foundations: complete attribution, use-case-driven copy, active Q&A, Pro-grade metadata, and channel-specific content. The tools just make that work visible faster.
That’s the work we do every day across Amazon, Walmart, Home Depot, and the rest of the modern retail stack. And then some.
About And Then Some Marketing: ATS is an omnichannel retail performance partner for consumer goods brands, with 20+ years of experience and $2B+ in retail sales supported across e-commerce, marketplaces, and brick-and-mortar. Talk to our team →
