Measurement
Part of Why talent agencies measurement and reporting needs a 2027 rethink
When to rely on talent agencies benchmark data
A practical guide to talent agencies benchmark research, showing how to score source quality, market sizing and disclosure evidence before you rely on a number.
What to take away
- Decide first whether a benchmark is meant to size a market, judge a peer group or evidence a disclosure claim. The three need different sources.
- Published sector statistics rarely cover talent management on their own, so treat any single figure as a starting point and check the method behind it.
- Score every source on recency, sample, definition and whether the underlying rule still applies before it enters a board pack.
- If a benchmark cannot be traced to a named publisher and date, exclude it from decisions about pricing, hiring or client reporting.
What talent agencies benchmark research can and cannot settle
Market sizing and the limits of published data
Market size questions are the hardest to answer from published material. The Office for National Statistics publishes business statistics covering counts, turnover and business demography, but its industry groupings rarely isolate creator representation as a standalone category. Read the classification notes before you quote a figure. A number that covers all of advertising, for example, tells you little about a ten-person agency in Manchester.
Where you cannot find a clean category, use a labelled illustrative example instead of a borrowed statistic. If a comparable agency reports average retainer income of £3,500 a month, present it as an example drawn from your own sample, not as a market fact. The measurement and reporting guide sets out how to present sampled figures without overstating what they show.
Peer comparison and disclosure evidence
Peer benchmarks answer a different question: how does this roster, rate card or campaign perform against similar agencies? That requires a defined peer set and a consistent period. Disclosure benchmarks are narrower still. The Advertising Standards Authority's advertising codes set the recognition rules that influencer marketing must meet, so any benchmark claiming to measure disclosure rates should state how compliance was judged against those rules.
This is where most benchmark research quietly fails. A disclosure percentage drawn from a small convenience sample, judged against no published standard, cannot support a claim in a client report. Before you cite one, ask which rule or definition the researcher applied. If the answer is vague, drop the figure. The common measurement mistakes in England tend to start exactly here, with a number that was never defined tightly enough to be checked.
How to score a benchmark before you rely on it
The rubric
Use a fixed rubric so different team members reach the same verdict. Score each source out of 20 and record the reasoning alongside the score.
Show the numbers
| Publisher and date | 0–5 |
|---|---|
| Sample and method | 0–5 |
| Definition fit | 0–5 |
| Regulatory fit | 0–5 |
A source scoring below 12 should be cited only as context, never as the basis for a pricing or hiring decision. Record the score in the same spreadsheet as the figure, so the caveat travels with the number.
Turning scores into a reporting line
Once scored, decide how each benchmark enters reporting. High-scoring sources can sit in client-facing decks with the publisher and date named in the sentence. Mid-scoring sources belong in internal planning notes with the limitation stated. Low-scoring sources should be dropped entirely, however convenient they are.
Regulatory context changes the calculation. The Competition and Markets Authority publishes its activity on GOV.UK, covering competition and consumer enforcement that can affect creator marketing arrangements. A benchmark built before a significant enforcement action may no longer describe the market you are trading in.
Where benchmarks go wrong in practice
Mixing market size with agency performance
The most common error is treating a market sizing figure as evidence of agency performance. They are not interchangeable. A growing sector does not mean a given roster is growing, and a flat sector does not mean a well-run agency is struggling. Keep the two in separate sections of any report, with the source and date attached to each.
When you cannot source a defensible market figure, say so plainly. "No reliable published estimate covers this segment" is a stronger line than a borrowed number that a client can dismantle in one question.
Refresh cycles and version control
Benchmarks decay. Set a review date when a figure enters a report, and check the ONS business statistics when a new release covers your sector. Keep one master list of sources with scores, dates and owners. Without version control, two decks will quote different figures for the same claim.
Common questions
How often should benchmark sources be refreshed?
Review client-facing figures at least every 12 months, and sooner if a regulator or statistics body publishes a relevant update. Internal planning figures can follow the same cycle.
Can I use a competitor's published rate as a benchmark?
Only as a single data point with its source and date named. One published rate is not a market benchmark, and it should never be presented as an average.
What if no reliable benchmark exists for my niche?
Run your own small sample and label it clearly as an illustrative example. State the sample size, period and method so readers can judge it.
Should disclosure benchmarks reference the CAP Code?
Yes, if the benchmark measures marketing recognition. The ASA's advertising codes are the standard against which disclosure practice is judged in the UK.
