How Creatrbase vets creators
The method, in plain language. What we check, how we score, what the confidence marks mean, and how a verdict is reached.
What Creatrbase is
Creatrbase is an independent creator-vetting service for agencies. You bring a brief or a list of creators; we produce research on each one — modelled estimates, delivery confidence figures, a risk register, and a verdict — so you can decide who to work with. The service replaces the research portion of the work, not the judgement of the people who commission it. Every figure is a modelled estimate carrying a stated error band, and every finding is set out with the evidence and the method behind it. We do not promise campaign outcomes; we set out what we checked, how we checked it, and what we found.
Why this exists
Creator discovery is still, at most agencies, a manual job. Hours of scrolling, spreadsheets, screenshots, gut feel — repeated for every brief, under deadline. Doing it properly needs data access that platforms meter and price for software companies, not for a shortlist of fifty. So agencies pay twice: once in hours, once in subscriptions to creator databases that return profiles, not decisions.
We built the pipeline once so you don’t have to. Creatrbase points it at what brands actually evaluate — modelled delivery against your targets, brand safety on the record, audience fit with the working shown — and delivers the finished judgement, checked the same way for every creator, every time.
What we check
For each creator, the dossier covers:
- Attested metrics. Audience size, true reach, median views, engagement rate and posting cadence — each with its source and the date it was read.
- Engagement quality. Whether the engagement is consistent with the audience size, and how it holds across recent output.
- Audience composition. Where the audience is, and how much of it overlaps with your existing reach, which drives the overlap discount.
- Risk register. A content-history scan against your stated sensitivities, plus a platform-standing check. Findings are evidence-first: severity, the evidence with dates, the method, and the quantified effect.
Where a figure could not be obtained, the dossier says so plainly and names the reason — never a blank, never an invented value. Absence of findings is expressed as what was checked and found clear.
How scoring works
Expected delivery is modelled against the targets in your brief and expressed as a range with a central estimate, not a single number. From that we derive a hit probability — the modelled probability that a creator clears your target — which always carries an error band. The working sits beneath each derived figure: the inputs, the named derivations, their confidences, and the two or three heaviest assumptions. You can see how every number was reached.
The acceptance line is the probability at or above which we mark a creator as clearing your target. It defaults to 75% and is adjustable per brief. On the shortlist cover, the line is drawn visibly across the ranked creators; where fewer clear it than your brief asked for, the shortfall is stated in words on the cover, never hidden.
What the confidence tiers mean
Every figure carries one of three confidence marks, with its source and date adjacent:
- High confidence speaks plainly — the figure is well-sourced and we state it directly.
- Medium confidence is directional, and we show the basis for it.
- Low confidence is explicitly an estimate. Low confidence looks honest, not dangerous; it tells you where to press, not that the figure is worthless.
How a verdict is reached
Each dossier ends with one of three verdicts, with the reasoning immediately beneath it:
- Proceed. The creator clears the brief and the risk register is clean or minor.
- Proceed with cautions. The creator is workable, but there are findings you should brief around; the cautions are named with their quantified effect.
- Do not proceed. A finding or a shortfall makes the creator a poor fit for this brief; the reason is stated.
Independence
A creator’s purchase of any Creatrbase product never influences an agency-side vetting outcome. One engine, two sides of the table, no pay-to-play.
Questions agencies ask
What is Creatrbase for agencies?
Creatrbase is a creator-vetting service for agencies. You submit a campaign brief; we deliver a dossier per creator — modelled delivery against your targets, a brand-safety record, audience fit, and a verdict — with the working shown, within 48 hours. Every dossier is reviewed by a person before release.
How is Creatrbase different from creator databases like Modash or Kolsquare?
Databases sell you search over creator profiles — you still do the judgement. Creatrbase delivers the judgement itself: a uniform, documented vetting method applied to every creator on your shortlist, ending in a verdict you can forward to your client.
What does a dossier contain?
Eight sections, in order: the methodology it was built by; the attested metrics with sources and dates; modelled delivery against your targets; the hit probability with its error band; audience composition and overlap; the risk register with evidence; the working shown beneath every figure; and the verdict, last.
How fast is delivery?
Five full dossiers, or the shortlist for a paid brief, inside 48 hours of an accepted brief. A person reviews every dossier before release, and the clock pauses whenever a clarifying query is with you — it restarts when you answer.
What does it cost?
Brand Safety Scan is £39 per creator; Score is £7 per creator (minimum 10); Vetting Batch is £15 per creator (minimum 10); the Full Brief is £695 for up to 50 creators (£895 across three or more platforms); retainers run from £750 to £2,850 a month. Every organisation gets one free run: your real brief, five full dossiers, no charge. Prices exclude VAT.
See it for yourself
The clearest account of the method is a dossier. Read the sample dossier, or bring us a brief and see five real ones inside 48 hours.