Buy ReFiBuy if the goal is winning ChatGPT and Gemini product cards; buy Anglera if the goal is a catalog that is correct at the source. They overlap on enrichment but optimize toward different endpoints — many teams run both.
Both claim to enrich product data. This page is about where that claim stops.
ReFiBuy is built to make SKUs legible to AI shopping agents and to keep scoring them as models change; the question this page asks is a different one — what governs the values that get written, and who owns them when the next model reads the SKU differently.
Mine every spec from every source.
Every value traced to a document you can open. The catalog is only as honest as what it was built from.
Aim the catalog at the buyer who actually buys.
Grounded data still loses if it answers questions nobody asked. Alignment is what turns specs into conversion.
Product data is a practice, not a project.
Markets move, suppliers reissue, buyers change what they ask for. A catalog that is right in March is wrong by August unless something is watching.
Scored against public documentation. Grouped by the three acts — so you can see which ones ReFiBuy leaves on your desk.
Catalog feeds, site pages, reviews, Reddit; no supplier documents
PDFs, spec tables, drawings, manuals, images, sites
Generates attributes to meet engine specs, not new schema fields
Proposes fields your schema never had
Normalizes values at SKU level; no versioned pick lists documented
Normalizes and governs allowed values, versioned
Maps SKUs to AI engine product cards; hierarchy classification unclear
Auto-classifies; channel and marketplace mapping
Each recommendation backed by citations and confidence scoring
Every value cites its source doc and page
Vertical and use-case context; no B2B versus B2C variants
B2B specifier and B2C shopper enriched differently
Reviews, Reddit mentions, competitor standards feed the enrich loop
Reviews, search, competitor rails, social — fed back
Titles, descriptions, bullets, key features, and Q&A pairs
Original copy per persona and channel
No image generation found in public material
Generates studio-grade imagery for photoless SKUs
Closed-loop monitoring, recurring jobs, re-enriches as engines evolve
Re-enriches on its own after go-live
Eligibility scorecard on four dimensions, tracked over time
Scored against your standards; nothing publishes below bar
Syncs to PIM and ERP; named connectors still early access
Writes back to PIM, ERP, warehouse, commerce
REST API, webhooks, and MCP server; docs gated in early access
API, webhooks, and MCP servers
Software with approval workflows; full-automation mode offered
Anglera owns the work; review is a guardrail
“Buyer signals” is the emptiest phrase in this category, so here is the literal thing. Each of these is an observation from a live market, the gap it exposes, and the field that gets created as a result.
"nema 4x enclosure 12x10x6 hinged" returns zero results, repeatedly, from buyers who then bounce — while the catalog carries dozens of enclosures that match
The environmental rating lives in description prose ("suitable for washdown and outdoor use"), so it is readable but not filterable and not answerable
enclosure_environmental_ratingNEMA 1 | NEMA 3R | NEMA 4 | NEMA 4X | NEMA 12 | NEMA 13 | IP54 | IP65 | IP66 | IP67"Listing says stainless. It's 430 — the bracket pitted within one season 200 yards from the water. 316 or don't call it marine."
Material is modeled as one bucket, "stainless steel", so grade — the part that predicts corrosion behavior — never reaches the page or the agent answering "will this rust near salt water"
material_alloy_gradeAISI 304 | AISI 304L | AISI 316 | AISI 316L | AISI 430 | 17-4 PH | UnspecifiedSeveral hundred drivers flagged for a missing required attribute; the dimming method is present, but only as a title suffix — "…-010V", "…TR", "…D2" — parsed off the SKU by whoever built the feed
Dimming protocol was never a field, so it is not validated, filtered, or asserted anywhere; each channel re-derives it from a naming convention that predates half the SKUs
dimming_protocol0-10V | 1-10V | TRIAC (forward phase) | ELV (reverse phase) | DALI-2 | PWM | Non-dimmableReFiBuy scores every SKU on four stated dimensions — content quality, crawlability, semantic quality, contextual signals — then generates and enriches to close gaps, with an Enrichment Queue putting human review before publish. It is built to make a catalog legible to AI shopping agents. Anglera's argument is about what a legibility score measures by construction: all four dimensions are properties of the text. A clean title reads well whether the thread underneath is BSPT or NPT — only the datasheet settles that. Correctness moves at the speed a person reads a PDF. And where free text lands instead of a governed value, "black oxide", "blackened" and "BLK OX" arrive as three strings for one finish, and the filters return nothing. Anglera runs it the other way: fix the attribute and its value set, source every value to a document, then let anything downstream score it.
Values are normalized into a governed, versioned set of allowed values — so a filter works, and keeps working after the next import.
3/4 in0.75"3/4"19mm3/4 inchDN200.75 in (DN20)Six suppliers, six spellings, one physical size. Filters only work once they agree.
BlkblackBLACK MATTEMatte BlkRAL 9005Black — MatteFree text makes a colour filter useless. A governed value makes it a facet.
SS316316 StainlessStainless Steel 316A4 StainlessStainless Steel — 316 / A4Same alloy, four vocabularies, plus a trade name. Buyers search all of them.
Filling the fields you defined has an invisible ceiling: a catalog can hit 100% complete and still miss the attribute that loses the sale, because completeness is measured against a schema someone drew years ago. Schema Foundry reads competitor listings, buyer searches, review complaints and your supplier docs, and proposes the fields you never defined — which is where ReFiBuy stops.
How Schema Foundry works
ReFiBuy sells an "Agentic Commerce Optimization" platform built on a Commerce Intelligence Engine that runs a six-step loop: ingest, evaluate, enrich, distribute, sync, monitor. It scores every SKU across content quality, crawlability, semantic quality, and contextual signals, then generates titles, descriptions, bullets, and Q&A pairs so products are legible to AI shopping engines like ChatGPT, Perplexity, Gemini, and Claude. It distributes to OpenAI's Agentic Commerce Protocol and Google's UCP, and syncs enriched data back to systems like PIMs and ERPs. Founded 2025 by ChannelAdvisor's Scot Wingo; raised a $13.6M seed in May 2026.
Pricing: Not publicly disclosed — there is no pricing page and no listed tiers or dollar amounts. Their developer page states "SKU-based pricing you can forecast," and public material describes an annual subscription scoped by catalog size and monitored SKUs. Access currently runs through a design partner program and early access, with broader availability signaled for Q3 2026.
ReFiBuy websiteDTC and retail teams whose first priority is measurable visibility inside AI shopping engines, and who want SKU-level eligibility scoring against competitors on the offer card.
We'd rather tell you here than in month three of an implementation.
Capability verdicts reviewed against ReFiBuy's public documentation on July 14, 2026. Vendors ship quickly — if something here is out of date, tell us and we'll correct it.
Bring one category and your supplier files. In 30 minutes you'll see it enriched — complete, structured, and consistent enough to launch on — plus the attributes your schema didn't have yet.
Book a demo