Key Takeaways
- On June 30, 2026, Revmatics, an AI-driven commerce company, announced it had acquired DataFeedWatch, a leading product feed management platform, from Cart.com in a cash-and-stock deal.
- DataFeedWatch supplies the plumbing that distributes product data to 18,000+ shopping channels, while Revmatics' Lumara AI provides the intelligence layer that decides what to say, to whom, on which channel, and at what price.
- As AI systems discover and recommend products, the quality of a product feed (product data formatted for each channel) directly shapes visibility and conversion. Optimizing feeds for AI discovery is becoming as essential for merchants as search optimization once was.
Revmatics has acquired DataFeedWatch
The deal unites DataFeedWatch's product feed distribution across 18,000+ brands and thousands of shopping channels with Revmatics' Lumara.
www.newsfilecorp.comLehi, Utah-based Revmatics announced on June 30, 2026 that it had acquired the product feed management platform DataFeedWatch from Cart.com. The deal is structured as a combination of cash and stock, with Cart.com taking an early equity position in Revmatics as the seller. Revmatics CEO and cofounder Ricky Ray Butler frames the acquisition not as a feature add but as a statement about where commerce intelligence is headed.
Two terms are worth defining up front. The first is product feed, the data that describes a catalog's items (titles, prices, inventory, images, categories) formatted to match what each selling channel such as Google Shopping, Meta, or TikTok requires. The second is shopping channel, the umbrella term for the places where consumers can find and buy products, from search-driven ad surfaces and comparison sites to social storefronts. DataFeedWatch has served as the connective tissue between the two, trusted by brands for more than a decade.
The plumbing DataFeedWatch has provided
Tracing DataFeedWatch's origins gives this acquisition its depth. Founder Jacques van der Wilt launched the company in 2012 as a web-based tool for optimizing product data for Google Shopping and other comparison shopping channels. It later expanded to support 2,000-plus channels across more than 60 countries. In 2022, Cart.com acquired DataFeedWatch and grew it under its wing. The Revmatics deal marks another change of ownership from there.
The platform's reach today is substantial. It syndicates product data to 18,000-plus shopping channels including Google Shopping, Meta, Bing, and TikTok, and has managed close to a billion products. As the press release puts it, DataFeedWatch is in effect the connective tissue between a brand's catalog and every place a customer can find it, translating merchant product information into the exact shape each channel demands and delivering it there.
Those results show up in the numbers. Across 63 customer case studies cited by Revmatics, the following improvements were reported. They suggest that feed optimization is not mere format conversion but a step that shapes the quality of visibility and conversion.
| Metric | Improvement |
|---|---|
| Return on ad spend (ROAS) | 29-97% increase |
| Cost per acquisition (CPA) | 18-46% reduction |
| Conversion rate | 11-43% higher |
| Click-through rate (CTR) | 13-57% higher |
According to the company, these gains come from optimized product titles, accurate inventory data, sharper categorization, and AI that streamlines feed management end to end.
Layering on an intelligence layer called Lumara
Where DataFeedWatch handles distribution, Revmatics supplies the intelligence layer that sits on top. At its core is Lumara, a patent-pending, self-improving agentic AI. In Butler's framing, where DataFeedWatch distributes product data, Lumara decides what to say, to whom, on which channel, at what price, and how to convert. Combined, the two aim to optimize the entire revenue funnel, from audiences and messaging to distribution, highly personalized landing pages, and even real-time retail store performance tracking.
What the company repeatedly emphasizes about Lumara is a design philosophy that runs counter to the industry's brute-force approach of scaling ever-larger models. Revmatics positions Lumara as a "Level 4" self-improving AI that dynamically selects the right model for each task and eliminates bloat, delivering precise performance on a fraction of the energy and cost. Butler describes marketers as "the masters of waste reduction" and frames energy-efficient AI as a competitive weapon. Whether that claim holds in practice remains to be tested in real deployments, but it is worth understanding as the thinking behind the deal.
For DataFeedWatch's existing customers, the change is additive at first. Continued access to the feed management tools they rely on stays in place, now with the option to layer Revmatics' intelligence suite, enterprise capabilities, and coaching on top.
Why the product feed supply layer is strategic ground in AI commerce
Viewed only as a single-company M&A, this deal misses its point. The larger context is a shift in which the product data supply layer is becoming strategic ground as AI systems discover and recommend products.
In the 2026 commerce landscape, AI agents are beginning in earnest to discover, evaluate, compare, and purchase on behalf of consumers. According to Shopify data, AI-driven traffic to its stores grew roughly 8x year over year in Q1 2026, while orders from AI-powered searches rose nearly 13x. Products recommended by AI are selling without a shopper ever viewing a product page, and that kind of purchase is starting to make up a real share of activity.
This changes the standing of product data at a fundamental level. The website used to be the primary customer touchpoint, but in the age of AI shopping, the product feed that generative engines such as ChatGPT, Perplexity, Amazon Rufus, and Google Gemini consult becomes the front line. Here is where GEO (Generative Engine Optimization) enters. GEO is the practice of structuring product data and brand content so AI engines can discover, understand, and recommend products correctly. If SEO optimizes for search engines, GEO optimizes for generative AI.
Revmatics reaching for DataFeedWatch reads as an effort to own precisely this supply layer. For an AI agent to recommend the right product, inventory, pricing, category, and title all have to be accurately structured and delivered to each channel beforehand. No matter how smart the intelligence layer (Lumara) is, thin data in the supply layer caps the quality of recommendations. Put the other way around, if you can place AI judgment on top of infrastructure that already flows product data to 18,000-plus channels, those improvements can ripple across the whole network, in the company's words, "overnight."
How merchants should organize their product data
So what should real merchants and retailers prepare for? The implications are relatively clear.
First, treat the product feed as the primary customer touchpoint rather than supporting data for ads. Feeds deserve the same continuous attention once paid to the website. Reflecting actual stock levels rather than approximate availability, making pricing signals transparent, and shoring up review and trust signals become baseline conditions for being selected by AI agents. The industry is beginning to describe frameworks that score "agentic readiness" across several dimensions, including structured data quality, machine-readable product feeds, transparent pricing signals, review and trust signals, brand authority, and API accessibility.
Second, plan for the reality that multiple AI shopping engines coexist. This is no longer an era where a single formatted feed suffices, because each engine demands different data shapes and weighs different factors. A feed management foundation like DataFeedWatch earns its value by automating this many-to-many translation and absorbing channel-by-channel differences.
Third, the choice of where to entrust this supply layer will increasingly become a vendor-selection question. Some players, like Revmatics, aim to vertically integrate the feed and the AI intelligence layer, while others move to connect openly along standard specifications. How much you let your product data depend on whose infrastructure is a decision that warrants careful judgment.
Conclusion
Revmatics' acquisition of DataFeedWatch is a symbolic move showing how strategic an asset the product feed supply layer is becoming in an age where AI discovers products. Layering Lumara's judgment on top of DataFeedWatch's distribution plumbing targets a simple truth: the quality of AI recommendations ultimately rests on the quality of the underlying data.
The lesson for merchants is plain. The foundation for getting an AI agent to pick your products lies not in a clever algorithm but in accurate, structured product data itself. Optimizing feeds for AI discovery is shifting from an experiment run by a few advanced companies into a routine operational task that shapes visibility and sales. What to watch next is how quickly Lumara's intelligence actually reaches the 18,000-plus brand network and what standard playbooks it brings to the practice of GEO.
