We understand your products better than you do.
From a title and a GTIN, Central produces 50+ verified specs, channel-perfect content, and a confidence score on every fact. Automatically.
See a sample knowledge cardThree ways in. One engine.
All three paths land at the same enrichment workflow.
CSV import
Upload your catalog. AI maps your columns to product fields. Variants and custom fields handled.
URL list
Paste up to 50 product URLs. Central reads the pages, creates the records.
Website discovery
Paste your shop URL. Central crawls representative categories and products.
All three land at the same engine.
Verified specs. Source-attributed. Confidence-scored.
Real product, real verified data. Not a mockup.
Apple iPhone 17 Pro Max
256GB · Natural Titanium
| Attribute | Value | Sources | Conf. |
|---|---|---|---|
| Weight | 219 g | 14 | 0.94 |
| Battery (video) | 36h | 12 | 0.91 |
| Camera zoom | 10× optical | 11 | 0.88 |
| Display peak | 2,500 nits | 9 | 0.85 |
Four rules. No exceptions.
Two sources or it didn't happen.
Every published value has independent agreement across multiple sources. If the engine can't find corroboration, the value stays hidden — not invented.
A confidence score on every fact.
Every value carries a score and a source list. Sort by it. Filter by it. Audit by it.
Anti-hallucination is built in.
Generated content is checked against verified facts before it goes anywhere. Numeric mismatches, invented certifications, unsupported claims — all caught before publish.
Your data wins ties.
Merchant and brand data outranks third-party. The engine never overwrites your inputs without your approval.
gather. enrich. sell.
Read 15+ sources per product — manufacturer pages, retailer listings, spec sheets, reviews, regulatory databases.
Extract candidate values, normalize across units and languages (1,640 g = 1.64 kg), consolidate into one confident answer per attribute.
Reshape verified canonical content for every channel — Google, Amazon, ChatGPT, Bing, Perplexity, your own product page.
Same engine. Category-aware extraction.
Knowledge Graph mapping to ~6,000 Google Product Categories means you don't pick a taxonomy; the engine picks the right one.
Things people ask before they paste a URL.
How does Central know which sources to trust?
Independent sources, multi-source agreement, brand and merchant authority, source quality history. If one third-party page disagrees with five others, the five win.
What happens to my own data?
It wins. Central never overwrites merchant or brand data without your approval — your data is the canonical source.
What about fields where Central can't find verification?
Stays hidden. We'd rather show fewer fields with confidence than more fields with guesswork.
Can the engine run on a niche category?
Yes. The architecture is category-agnostic — the engine adapts field suggestions, extraction, and optimization to whatever category your catalog is in.
How long does first enrichment take?
First product within 5 minutes. A full catalog runs overnight depending on size.