Audience research
Part of Audience research for property creators with populations and permissions kept apart
Auditing property creator competitors without inferring anyone's market share
Audit property creator competitors in England without inferring market share by separating entity, service, review, visibility and regulatory evidence.
Competitor research should produce a dated evidence file. A company registration, website claim, customer review and search result answer different questions. None establishes England market share, service quality or commercial demand.
What to take away
- Competitor research should produce a dated evidence file, not a league table.
- A registration, website claim, review and search result answer different questions.
- No single source establishes market share, service quality or commercial demand.
- Keep output unranked unless a complete population and defensible measure exist.
- Treat visibility as a snapshot that varies by time, location, device and personalisation.
Set the inclusion rule
- Define the service, English geography, customer type and research date before searching.
- Include a business only when a current first-party page explicitly offers the in-scope service to the chosen customer or area.
- Exclude directories, duplicate trading names, dormant pages, unrelated estate agents and creators whose commercial service cannot be verified.
- Keep the output unranked unless a documented, complete population and defensible ranking measure exist.
Verify identity separately
- Record the displayed trading name, legal name, company number, website and lookup date.
- Use Companies House company information to inspect incorporation, officers, filings and previous names. Treat filed records as identity evidence, not proof of quality, solvency or website ownership.
- For example, a Companies House search on a trading name returns a company number, incorporation date, officers and filing history, all tied to the lookup date.
- Match the company number shown on the first-party site. Mark discrepancies as unresolved rather than choosing the convenient identity.
Capture service claims without adopting them
- Save the exact first-party page, access date, wording, geography and exclusions for each service claim.
- Separate commissioned production, sponsorship, lead introduction, estate-agency activity and mortgage or investment promotion.
- Where a business may conduct residential estate or letting agency work in England, check the applicable redress requirement and the scheme's own live member record for The Property Ombudsman or the Property Redress Scheme. Membership is not an endorsement.
- For mortgage or other regulated financial services, follow the FCA's firm-checking process and record the precise permission and date. Authorised status neither validates an advert nor proves a service result.
Keep reviews in their own column
- Record platform, displayed count, rating, date range, sorting method and whether incentives are disclosed.
- Do not quote a reviewer without checking publication and copyright implications, and never turn a public review into a fabricated customer case.
- Note that the CMA's online consumer reviews work addresses fake, concealed incentivised and misleadingly presented reviews. A review-platform total is therefore neither a verified client count nor a representative satisfaction rate.
- Search the ASA's published rulings by legal and trading names. A ruling is an advertising record about the cited communication, not a score for the whole business.
Treat visibility as a snapshot
- Predefine each query and record date, time, device, language, signed-in state and stated location.
- Save results by page typeorganic website, map, advert, video or directory. Do not add unlike placements into a single rank.
- Google's first-party explanation says search results can differ with time, context, location, language, device and personalisation. One researcher's result cannot establish persistent visibility or market share.
Close with evidence gaps
For every business, report verified identity, attributed service claims, separate regulatory checks, review evidence, search snapshot and unknowns. Have a second researcher reproduce a sample and log disagreements. Remove personal contact details that are not needed, and do not record residents, properties or staff merely because they appear in public media.
A completed row uses the same fields for every business. The placeholder below shows the output format, with illustrative values only.
Field / Illustrative entry
- Trading name
- Example Creators, a placeholder and not a real firm
- Company number
- 00000000, placeholder number, from a dated Companies House lookup
- Service claim
- Lettings marketing, from a first-party page saved with its access date
- Redress
- The Property Ombudsman member record, checked and dated
- Reviews
- Count and rating, named platform, date range, sorting method, disclosure
- Search snapshot
- One query, desktop, signed out, stated location England, stamped with date and time
- Unknowns
- Website ownership, client numbers, satisfaction rate
Use the audit to shape a research question, not to claim superiority. A possible next step is interviewing eligible buyers about an unmet workflow, with voluntary participation and separate marketing permission. Stop if the legal entity is unclear, the in-scope service is not evidenced or the proposed comparison would require invented performance, price or customer outcomes.
Before you act
- Freeze the inclusion rule, identity fields and claim log before searching.
- Run the Companies House, redress and FCA checks as separate dated records.
- Keep reviews and search snapshots in their own columns, unranked.
- Arrange an independent second pass over a sample of completed rows.
Common questions
Why should competitor research not be presented as a league table?
The article says research should produce a dated evidence file, not a league table. A registration, website claim, review and search result answer different questions, and none establishes England market share, service quality or commercial demand.
How should company registration records be used in this audit?
Use Companies House to inspect incorporation, officers, filings and previous names. Treat filed records as identity evidence, not proof of quality, solvency or website ownership. Match the company number shown on the first-party site and mark discrepancies as unresolved.
What limits does a search result snapshot have?
Google says search results can differ with time, context, location, language, device and personalisation. One researcher's result cannot establish persistent visibility or market share. Save results by page type and do not add unlike placements into a single rank.
