What we analyse

Siguro scores a listing from seven independent signals. Every listing starts at 100, and each signal that fires subtracts its own fixed weight, set before we saw any listing at all. A low score means several signals came back badly at once — not that the seller is definitely a scammer.

1. Reverse image search

   Weight 30

   We look for the same photos elsewhere on the web. Pictures stolen from property listings in other countries are the strongest fraud signal we can measure, which is why this carries the most weight.

2. Account age

   Weight 18

   Accounts created a few days ago carry far more risk than accounts with a history. Scammers lose accounts often and make new ones.

3. Contact method

   Weight 18

   We check whether the account avoids phone calls and accepts in-app messages only. A visible phone number is harder to use anonymously.

4. Account name pattern

   Weight 10

   Names like “user48213773” are generated automatically at sign-up. A real name proves nothing, but a generated one marks an account made in a hurry.

5. Price vs. area median

   Weight 10

   We compare the price against the area median for the same kind of property. A price far below market is the classic bait: it looks like a find and rushes the decision.

   This signal is not active yet: we don't have enough data on area prices to compute an honest median. Until then it does not affect the score.

6. Historical account activity

   Weight 10

   We check whether other listings from the same account have already been flagged. A repeated pattern says more than a single listing.

7. Description anomalies

   Weight 4

   The listing text is read for phrasing typical of fraud: time pressure, deposits demanded before a viewing, refusal to meet in person.