What is Claro?
Claro is the operating system for product and supplier data. At its core is a canonical entity layer: it ingests supplier feeds and documents, resolves what is actually the same product, fills and validates attributes, and keeps ERP, PIM and downstream systems in sync as your catalog changes.
How is Claro different from a PIM?
A PIM stores and publishes product data you already have. Claro is the layer upstream of it — it decides what a product's canonical record should be from messy supplier inputs, before it ever reaches your PIM.
How is Claro different from a catalog enrichment tool?
Enrichment tools fill fields. Claro also resolves identity and duplicates, validates against your rules, and keeps all of it synchronized continuously — enrichment is one part of the workflow, not the whole of it.
How is Claro different from a data services company or managed service?
A managed service cleans your data once, by hand. Claro is software: a repeatable, confidence-scored pipeline you can run again on the next supplier file without paying for another manual pass.
What is the canonical entity layer?
The single, trusted representation of each real-world product — resolved from every supplier's version of it — that every downstream system (ERP, PIM, search, pricing) can reference instead of maintaining its own conflicting copy.
How does Claro keep data accurate as the catalog changes?
Claro re-runs matching, validation and monitoring continuously as new supplier feeds and price lists arrive, rather than treating the catalog as a one-time cleanup project.
Is Claro an AI tool?
AI does the reading, matching and drafting. Confidence scoring, provenance and rule-based validation decide what is trusted enough to write back automatically, and what goes to a human first.
What does a pilot look like?
A scoped engagement on one supplier flow or one category: a working pilot on your real files in about eleven business days, with measurable outcomes agreed up front before any decision to expand.
How long does it take to start seeing results?
Most pilots show a first working result — a mapped supplier file, a set of matched SKUs — within the first five days.
Does Claro require us to move to a new system first?
No. Claro sits above your existing ERP, PIM or commerce stack and writes back into it — it is not a system you migrate to.
What systems does Claro integrate with?
Common ERP and PIM systems, spreadsheets and CSV/EDI feeds, and commerce platforms such as Shopify, via API, file exchange or direct write-back, depending on what your stack already uses.
How does Claro handle products without GTIN, UPC, or shared identifiers?
Matching falls back to manufacturer and part-number combinations, descriptions, category-specific rules and semantic and attribute-level evidence — the same approach used for own-label and exclusive SKUs.
Is our data secure?
Your catalog and supplier data are processed under contract for your use only, with access controls and an audit trail on every change Claro proposes or writes back.
How is Claro priced?
Pricing is scoped around your workflow, catalog complexity, volume and the systems involved — most teams start with a pilot on one workflow, then move to continuous operations or an embedded/API deployment. The self-serve Research Agent has its own separate pricing. See pricing.
Do we need clean data before we start?
No — the messy supplier files are the input, not a prerequisite. Claro is built to take listino PDFs, inconsistent spreadsheets, BMEcat and EDI feeds exactly as they arrive and resolve them into canonical records.
What file formats and feeds can Claro ingest?
Excel and CSV, PDF price lists and technical documents, BMEcat and other industry XML, EDI feeds, and direct API or database connections — the same supplier gets handled the same way whichever format they send.
Can Claro classify to ETIM, eCl@ss or our own taxonomy?
Yes. Claro classifies to standard schemes like ETIM and eCl@ss, or to a taxonomy you define, and keeps the mapping consistent as new products arrive.
What happens to low-confidence matches?
Every row carries a confidence score, but confidence alone never decides what happens next. Your rules determine what's eligible to write back, the confidence score measures how certain the match is, and impact determines whether it needs a reviewer's sign-off. Trusted, low-impact matches can write back automatically; anything uncertain or high-impact is routed to a review queue with the evidence attached, so a person decides.
Who stays in control of what gets written back?
You do. You set the rules, the confidence thresholds and the impact tiers that decide what needs review and which fields can auto-write. Claro proposes; your rules and reviewers decide what reaches your systems.
Can Claro run inside our own product or systems?
Yes — beyond the Claro platform, the same engine is available as an API, embedded behind your product, or over MCP so AI agents can call it. See the platform page for the delivery modes.