Measurement
Part of UGC marketplaces measurement: keeping six evidence classes apart
UGC marketplaces benchmark research in England: no official figure to cite
Benchmark research for UGC marketplaces in England: no official body publishes a benchmark figure, so screen candidate sources and record the gap.
Benchmark research for UGC marketplaces in England begins with an awkward fact: no official statistics body in the United Kingdom publishes a benchmark figure for creator-marketplace activity, creator earnings or campaign rates. This article sets out where to look instead, which named sources can carry weight, how to screen a candidate against eleven admission fields, and the written criteria for keeping the evidence gap open rather than adopting a number that does not describe the asset in front of you.
It is written for the England buyer who has to decide something specific — repeat a commission, change a brief, or approve a supplier — and who needs to know what can honestly be cited in that decision.
What a benchmark has to be, and what does not exist
A benchmark is a published figure with a stated population, denominator, period and method, against which your own asset can be compared. In England, no such figure exists for UGC marketplaces from an official source. The Office for National Statistics publishes official statistics on employment, prices and business activity; its Annual Population Survey covers creative occupations, which is employment data, not marketplace or creator-earnings data. Commercial rate guides do circulate, but their publishers are market participants rather than statistical authorities, and their figures describe their own users, not a defined England population.
That is why a UK-wide, global or supplier average must never be relabelled as an England UGC-marketplace norm. The England market-size data and sources that do exist describe something different from a marketplace benchmark, and mixing them produces a figure nobody can defend.
Screen every candidate against eleven admission fields
Record every candidate source, accepted or rejected, against the same eleven fields:
- Source identifier.
- Publisher and publication date.
- Evidence class.
- Asset population and geography.
- Event definition.
- Numerator.
- Denominator or unit.
- Period and timezone.
- Method version.
- Exclusions.
- Reviewer.
Benchmark admission record fields
- Publisher, title, date, funding, conflicts
- Geography and collection locations
- Population, frame, eligibility, sample
- Definitionsmarketplace, creator, asset, campaign, outcome
- Platform, placement, format, product, paid status
- Numerator, denominator, identity unit, period, timezone
- Collection, coverage, weighting, uncertainty
- Privacy, visibility, permissions, recheck, withdrawal
If a material field is absent, mark it unknown instead of inferring a value. If the population, denominator or evidence class differs from your asset, reject quantitative comparison; a narrative source may still explain a method, but it cannot supply a target. Define the buyer decision and the metric before you search, so that a convenient number cannot redefine the question — the eligible asset population and metric definitions should be settled first.
Named sources worth checking in England — and what each can bear
Several named authorities are worth reading before you conclude that nothing exists, provided you are clear about what each one supplies.
- Office for National Statistics.Publishes official statistics and sets out quality dimensions such as relevance, accuracy, timeliness and comparability. Those questions can be applied voluntarily to your own record; doing so does not make your work official statistics.
- Office for Statistics Regulation.The Code of Practice for Statistics governs official statistics and voluntary adopters. The regulator's guidance Putting the Code principles into practice describes openness about quality and methods. It is a research discipline here, not a regulatory badge for a marketplace.
- HM Treasury.The Magenta Book, and in particular its Quality in Policy Impact Evaluation chapter, explains the standards an evaluation has to meet.
- Information Commissioner's Office.Guidance on data protection by design and default supports purpose limitation and data minimisation across a processing lifecycle.
None of these bodies publishes UGC-marketplace benchmark figures. They are still worth the reading, because each tells you what a defensible figure would have to contain before you could cite it.
Keep the evidence gap open on written criteria
An open gap is a decision, not a delay, so record it as one. Name the matter, the decision owner and the reviewer; state the review date; and set the triggers that would reopen the search — a publisher releasing fuller method notes, a revised denominator, a corrected table, or enough comparable, permissioned commissions for you to describe your own history. A changed source is re-admitted from the beginning rather than patched into an earlier decision.
Keep a rejection log of the searches that failed, with the source identifier, the fields inspected, the mismatch, the reviewer and the review date. Quote no proprietary dataset in it. Logs of this kind belong with your competitor research in England, so that a later recheck is reproducible.
Keep attribution, impact and personal data out of the benchmark column
A supplier-reported attributed event is not interchangeable with an outcome the buyer has reconciled independently, and neither is automatically a causal effect. Without a compatible counterfactual, no benchmark can validate incremental contribution or return on investment; the practical limits are set out in attribution methods for one asset. Creator, featured-person and audience records should be retained only where purpose limitation and data minimisation allow. Curiosity about benchmarks is not authority to hold or join those records.
