ipd-ranking-v1

Performance Score Methodology

The IPD Performance Score is a deterministic 0–100 organic discovery metric. It combines native published review performance with confidence weighting, deterministic profile completeness and server-controlled verification. Organic ranking cannot be purchased. The score is a shortlisting aid, not a guarantee of service quality or booking outcome.

Bayesian confidence, in plain language

A raw five-star average from one review is much less evidence than a 4.8-star average from twenty reviews. IPD therefore starts from the current site-wide native review mean and gradually lets a photographer’s own native published review history take over as more reviews accumulate.

weighted rating = (v / (v + 5)) × R + (5 / (v + 5)) × C

v = native published review count

R = photographer’s native published average

C = IPD-wide native review mean

5 = confidence prior used by ranking version 1

Native review component (61.111%)

The largest component uses only native published IPD reviews and confidence-adjusts their rating.

Published native reviews

Only reviews explicitly published through IPD moderation contribute. Pending, rejected, removed and legacy-visible reviews do not count.

Bayesian confidence

The native published average is blended with the site-wide native mean using a five-review prior, so one perfect review does not automatically outrank a substantial strong review history.

Profile and trust components (38.889%)

Only deterministic profile completeness and server-controlled verification contribute here.

Profile completeness (22.222%)

Ten deterministic public-profile checks measure whether useful identity, portfolio, service, location, experience and price information is present.

IPD Verification (16.667%)

Only IPD’s server-controlled Verified state contributes to this trust component.

Freshness (0%)

Freshness is excluded in ranking version 1 because the available update timestamp is photographer-editable. Its intended weight is redistributed across the safe components.

Editorial signals (0%)

Featured recognition, awards and other editorial placement may remain visible, but they never change the automatic organic score.

What the score does not use

Manual legacy rating/count fields have no direct effect. Neither do Featured recognition, awards, owner-editable timestamps, advertising spend, commercial relationships or a manually typed ranking value. If scores tie, native published review count and then stable profile ID ordering provide deterministic tie-breakers.

Use the comparison tool, full portfolios, pricing, availability and direct communication alongside this discovery signal. For the wider product principles, see Why IPD.