Trust for marketplaces

The problem

A new marketplace cold-starts with zero reputation signal. Every seller looks identical to every buyer, so the only differentiator left is price.

Incumbent platforms lock reviews inside their own walls. A seller with ten years of flawless delivery elsewhere starts at zero with you, which is exactly the seller you most want to attract.

Star ratings are also gameable and unauditable: nobody can check whether a five-star average was earned or manufactured.

The solution

• Read each seller’s portable reliability score directly: up to 100 sellers in one call, so a listings page costs a single request.
• The score is a deterministic function of an append-only ledger, and every point of it is explainable: you can show the factor breakdown, not just a number.
• Report your own outcomes back. A sale fulfilled on your marketplace becomes verified evidence in the same portable record, so participating makes the signal better for you too.

How it works

Then subscribe a webhook to score.band_changed and re-rank when a seller moves band, with no polling loop.

The API calls involved

• POST/api/v1/users/scoresBatch read, up to 100 external ids, partial success on unknown ids.
• GET/api/v1/users/{id}/score/explainFactor-level breakdown behind a seller’s number.
• POST/api/v1/eventsReport a fulfilled sale. Send an Idempotency-Key so retries cannot double-count.
• POST/api/v1/webhooksSubscribe to score.updated / score.band_changed and re-rank on change.
• POST/api/v1/users/{id}/share-tokenMint the token the embeddable badge resolves against.

$ curl -X POST https://api.credda.io/api/v1/users/scores \
    -H "Authorization: Bearer $CREDDA_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{"userIds":["seller_1","seller_2","seller_3"]}'

{ "scores": [ { "userId": "seller_1", "finalScore": 91.2, "scoreBand": "Excellent" },
              { "userId": "seller_3", "error": "not_found" } ], "count": 3 }

Full request and response shapes are in the API reference.

Start building