What counts as evidence?
Our initial sources are official pricing pages, terms, privacy notices and help articles. These document what a provider says. They do not prove response quality, security, reliability or the effectiveness of a deletion request. Each factual record carries sources, a check date and an evidence status.
Where official pages disagree, we report the conflict. Where a feature is not confirmed, we say that instead of treating it as absent. Hands-on labels will require an actual documented test with a date, plan, device and repeatable procedure.
How do we compare costs?
We separate monthly payment from annual monthly equivalent, identify currency and any observed tax caveat, and distinguish access fees from credits or add-ons. We do not compare a promotional annual price with an ordinary monthly subscription as if the commitment were the same.
A useful cost comparison includes the allowance and per-action tariff when available. If either is missing, we cannot honestly calculate a typical monthly usage bill. We will not fill the gap with an invented average.
What is the Heat score?
Heat scores documented capabilities out of five: what a platform lets you do, how you can shape the experience, and the available ways to explore it. It is not a hands-on quality test or a measure of sexual intensity. Every review names the use case being assessed and shows the reasons and sources behind its flames. Output quality, chemistry, value for money and security are not established by this score.
Three dimensions carry equal weight: Core capability, Choice & control, and Ways to explore. We assess each against its category and stated purpose. Character controls matter for a companion; room and session choices matter for cams; creator or item choices matter for fansites. A specialist does not lose points for omitting an unrelated format, and listing more formats does not prove they work better.
Each dimension earns 1–5 whole points. Three means the documented core workflow is supported with bounded choice; four adds meaningful optional controls or access paths; five requires multiple substantive controls or paths with their operation documented. Two means a major documented limitation, and one means a severely constrained workflow. Insufficient evidence is unassessed, not a penalty or a guessed score. These are editorial judgments, not scientific measurements or customer votes.
We average the assessed dimensions and round once to one decimal. At least two of three must be assessable. An unknown is excluded, not scored as zero, and incomplete coverage is labelled beside the score. Compare within category and inspect the breakdown: a higher number with a missing dimension does not establish superiority. Limited profiles remain unrated, not zero-rated.
Buying terms sit alongside the capability score, not inside it. Costs, cancellation, data handling and material policy conflicts remain in the review and its Before you pay panel; they can still change whether we recommend buying. A high Heat score is not a clean bill of health. Commission rates and sponsorship do not earn flames. Commercial opportunities may affect which clearly identified placements we develop, never these points or the matcher's results.
Rubric v2 replaced the earlier buying-terms Heat rubric on 27 September 2026. The old assessments concerned offer clarity, cost clarity, billing control and data control. Their findings remain visible separately; the new numbers measure something different, not an improvement in the products or evidence of a new hands-on test. Source dates remain distinct from the editorial assessment date.
What do the five flames mean?
In the matcher, the flames are labelled Match fit, separate from Heat score. They show the proportion of your selected features documented in a provider’s sources, with the exact count beside them. Five lit flames mean all your selected features are documented. A partial flame shows a partial unit: three documented features out of four fill three and three-quarter flames. No selected features means no meter.
This is not a five-star review, a measure of sexual intensity or a guarantee of quality, safety, privacy or enjoyment. A dim flame can reflect missing documentation rather than an unavailable feature. Open the provider’s review to inspect its sources and limitations.
The default directory is a browsing list, not a league table. Match results use selected documented features; commission rates are not an input. Source-based strengths and drawbacks remain editorial judgments rather than measured product performance.
How do corrections work?
Material changes should update the central provider record, affected review and comparison sections, and source history together. A new check date requires a real check. An old source is not quietly relabeled as fresh.
Spotted a mistake? Send the page, the specific claim and a current official source to the email on our Contact page. Keep account credentials, identity documents and intimate conversations out of your message.