Guru Score Methodology

The methodology can consider five categories of recorded evidence. Their relative weights are part of the product logic and may be adjusted when the scoring system is improved.

Evidence becomes less useful as it ages. A score may therefore decline when no fresh review, source update or useful feedback is available, even if the offer has not been explicitly reported as expired.

New evidence can also change the score in either direction. A corrected term, failed destination check, new community result or updated source record may alter the current evidence picture.

A numeric summary can hide important context, so Promocodes may also display evidence labels such as a stronger verified state or a limited-evidence state. These labels communicate what type of review was actually completed.

An offer with limited evidence can still be useful when the known facts are presented honestly. The objective is not to force every public listing into the same confidence category.

Shopper feedback can help identify changing conditions, but it is interpreted cautiously. A single success or failure may depend on country, customer type, account history, product selection, minimum spend, currency or timing.

Repeated feedback can become more informative when it is consistent with source data and recent review evidence. Abuse controls may also limit obviously repetitive or manipulated submissions.

The scoring system should operate on recorded evidence rather than invented values. Missing evidence remains missing, and a score should not be artificially improved by filling unknown fields with unsupported assumptions.

Administrators may correct factual errors in the underlying offer or evidence record. Such corrections should improve the source record rather than merely editing a number for presentation.

Use the score as one comparison signal. Read the recorded terms, review time, evidence label and destination details, then confirm the current conditions on the merchant site before paying.