Measurement
Part of Measuring property media with metric contracts, governed tracking and honest attribution
Why official records will not give you a property-media benchmark, and what will
Test whether England property-media benchmarks are defensible by checking geography, populations, definitions, denominators, periods and source limits.
This benchmark search for property creators found no defensible public England comparison for reach, engagement, enquiries, conversion, fees, returns or property outcomes. That is a bounded research result, not proof that no private dataset exists.
What to take away
- No defensible public England benchmark exists for property-media reach, engagement, enquiries, conversion, fees or returns.
- Official records such as ONS, HM Land Registry and Planning Data measure businesses, transactions or planning, not content performance.
- Build an internal comparator using one unchanged metric contract with defined events, denominators, periods and exclusions.
- An internal baseline describes prior performance under stated conditions and does not prove what would have happened without the media.
- Record the benchmark search as none defensible and re-run it when an official dataset changes scope.
Search method and scope
The editorial research team searched records available on 5 September 2026 from ONS, HM Land Registry, the England Planning Data service, the UK Statistics Authority and ASA/CAP. The target population was organisations or adults producing or commissioning media about identifiable property in England. Eligible benchmarks needed a documented event, population, denominator, England geography, period, collection method, exclusions and uncertainty.
We excluded supplier marketing, agency case studies, unattributed figures, paywalled data that could not be inspected, global platform aggregates, UK figures without a defensible England split and sources that measured property transactions or general internet use rather than property-media performance. No commercial analytics account was accessed, and no result was tested.
Why the official records do not supply the target benchmark
ONS UK business activity, size and location counts enterprises and local units using the Inter-Departmental Business Register and SIC classifications. Its accompanying quality information explains coverage of VAT or PAYE-registered units and available geographies. Neither record isolates property creators or reports content events, enquiries or returns.
HM Land Registry Price Paid Data concerns specified sales for value lodged in England and Wales. Recent data can be incomplete, and the publisher does not guarantee fitness for a user's purpose. Transaction counts and prices therefore cannot stand in for content performance or creator-caused outcomes.
The England Planning Data documentation covers planning and housing datasets whose coverage varies by area. It warns that an empty result does not prove no constraint exists. This is property evidence, not a creator-audience denominator.
The UK Statistics Authority's quality framework calls for suitable data and appropriate methods that meet user needs. It is an evaluation screen, not a property-content benchmark, and using it does not make private reporting an official statistic.
ASA/CAP's 2026 influencer-marketing research examines recognition of influencer advertising, principally on Instagram and TikTok, with its own disclosed method. It provides evidence about ad recognition, not an England property-media engagement, enquiry or conversion benchmark.
Build a valid internal comparator instead
Start with one unchanged metric contract. Define the event, eligible England property and content population, numerator, denominator, unit, Europe/London period, source export, query version, inclusions, exclusions, owner and uncertainty. Choose a historical cohort only if channel, commercial model, property type, distribution, measurement consent and data maturity are comparable.
Publish counts and missingness beside rates. Do not pool means with medians, views with people, sessions with enquiries, paid with unpaid distribution, or newly released assets with mature ones. If a material definition changes, break the series or restate it transparently.
An internal baseline describes prior performance under stated conditions; it does not show what would have happened without the media. The 2026 Magenta Book requires stronger evaluation reasoning for causal claims, including attention to a counterfactual and alternative explanations.
Record the benchmark search as "none defensible" until a source meets every criterion. Re-run it when an official dataset changes scope, a competent industry study publishes inspectable England methods or the intended population changes. Statistical, property, privacy, security, accessibility, advertising, FCA and rights reviewers remain separate gates even when a comparator is methodologically sound.
Before you act
- Define one unchanged metric contract before comparing any periods.
- Check channel, commercial model and property type are comparable.
- Publish counts and missingness beside every rate.
- Do not pool means with medians or views with people.
- Break or restate the series if a material definition changes.
- Record the benchmark search as none defensible until criteria are met.
Common questions
Why can HM Land Registry Price Paid Data not serve as a property-media benchmark?
It concerns specified sales for value lodged in England and Wales, and recent data can be incomplete. Transaction counts and prices cannot stand in for content performance or creator-caused outcomes, so it fails the benchmark criteria.
What must an eligible benchmark include?
It needs a documented event, population, denominator, England geography, period, collection method, exclusions and uncertainty. The search excluded supplier marketing, agency case studies, unattributed figures, paywalled data, global aggregates and UK figures without a defensible England split.
When should the benchmark search be re-run?
Re-run it when an official dataset changes scope, a competent industry study publishes inspectable England methods, or the intended population changes. Until a source meets every criterion, the result stays recorded as none defensible.
