How we verify reviews

What a review here is actually about

A review on this platform is about a commercial account, not about a person. The subject is the transaction: what was promised, what arrived, how long it took, how communication went, and whether the price matched the delivery.

Comments about someone’s body, appearance, identity, or private life are not reviews of a service. They are removed, and repeat offenders lose the ability to post.

Step 1 — the reviewer has to be a real, reachable person

• •Every review requires a confirmed email address. Unconfirmed submissions are never published.
• •One review per person, per creator. The database enforces this — it is not a policy we have to remember to apply.
• •Submissions are rate-limited by a hashed network address — we never store the address itself. Bulk submissions from one source are held automatically.
• •A review must be at least 40 characters of actual description. One-word drive-bys are rejected at the point of entry.

Step 2 — moderation before it goes live

Reviews are screened before publication, not after. The screen is for policy compliance, not for sentiment: a detailed one-star review that describes a genuine delivery failure passes. A five-star review that is really an advert does not.

• •Content about the person rather than the service — rejected.
• •Real names, locations, workplaces, or any attempt to identify someone behind a handle — rejected, and the account is banned.
• •Obvious competitor or agency posting — rejected as a conflict of interest.
• •Anything unlawful — rejected and, where required, reported.

Step 3 — disputes, and who carries the burden of proof

A creator can flag any review about their account. Flagging does not delete anything, and it never will. What it starts is a defined process.

When a flag disputes the facts, we ask the reviewer for evidence — a receipt, a transaction record, a message thread showing the promise. The burden sits with the person making the factual claim, not with the person disputing it.

If the reviewer produces nothing within the window, the factual claim does not stand and the review comes down. If they do produce it, the review stays, and the creator’s reply is published alongside it.

Step 4 — the right of reply

Any creator who verifies control of their account can reply publicly to any review of it. The reply sits directly under the review, at the same visual weight — not buried, not collapsed.

Verification is a code placed in the public bio of the account being claimed. That proves control of the account without asking anyone to send us identity documents, which we do not want to hold.

What we refuse to do

• •We do not sell review removal. There is no price, no package, and no private arrangement. The database has no removal reason that means "the subject asked" or "the subject paid" — the escape hatch does not exist in the schema.
• •We do not write reviews ourselves, and we never seeded this platform with invented ones. Editorial content on this site is written by us and labelled as such; reviews come from users or they do not exist.
• •We do not rank creators higher for paying us. Placement can be bought — it is called Featured, it is labelled everywhere it appears, it is shown outside every ranking, and it never changes a rating or a position.
• •We do not publish a subscriber star average until there are enough real published reviews behind it. Our rating, the editorial score out of 10, is a different thing: it is written from public facts, labelled as ours, and its method is published.

The moderation log

Every moderation decision is written to an append-only log, and the public entries are readable by anyone. The log records what was acted on, what the decision was, and why.

The log cannot be edited or deleted after the fact — that is enforced by the database, which rejects any attempt to update or delete a moderation entry. This is what makes the promises above checkable rather than merely stated.