Audience research
Part of Media buying agencies audience research for briefs
Researching rival media agencies without turning their claims into facts
Research media agency competitors using service pages, public records, ad libraries and trade directories without turning claims into facts.
Competitor research for a media agency should reveal who serves the same buying problem, how each firm describes its scope and what a prospect can verify. It should not copy case-study claims into a league table.
What to take away
- Competitor research should show who serves the same buying problem, not repeat case-study claims as facts.
- Record self-description, public record and independent research separately, and leave blanks rather than inferring missing detail.
- Certifications and directory membership are dated facts, not blanket quality marks for every team or channel.
- Public ad libraries and tender notices show visible activity but not agency involvement, spend or commercial performance.
- Refresh volatile fields and nominate an owner, because personnel, propositions and corporate records change.
Fix the comparison question
- Name the client type, channels, territory and service boundary under study.
- Decide whether the competitor set includes networks, independents, specialists and in-house consultancies.
- Set one research date for every entry.
- Write exclusion rules before looking at familiar names.
The IPA's member directory can identify member agencies and provides an agency-finding route. It represents IPA membership, not every provider operating in England, and membership itself is not a score.
Build an identity record
- Capture the exact trading and contracting names.
- Check registered status, officers and filings where relevant.
- Record office geography without treating a serviced address as delivery capacity.
- Note parent groups and related agencies disclosed by the provider.
Companies House explains the information and limits of its service in guidance on searching the register. A filing record confirms aspects of legal identity; it does not prove media skill, solvency or client satisfaction.
Build an identity record
- Capture exact trading and contracting names
- Check registered status, officers and filings
- Record office geography, not serviced address capacity
- Note parent groups and related agencies disclosed
Read service claims literally
- Save the precise service page, not just the homepage.
- Separate planning, buying, analytics, production and technology.
- Mark capabilities described for a global network but not clearly for the English office.
- Treat case studies and testimonials as first-party statements.
Do not infer platform access from a logo or award. Ask for proposal-level evidence if a capability affects eligibility.
Read service claims literally
- Save the precise service page, not homepage
- Separate planning, buying, analytics, production, technology
- Mark global capabilities not clearly for English office
- Treat case studies and testimonials as first-party statements
Inspect visible market activity
Meta's Ad Library allows searches for ads, with particular retention rules for some UK and regulated categories. It can show visible creative activity for named advertisers. It does not expose the agency behind every placement, the target audience, the full spend or commercial performance.
Public tender notices may also show how buyers describe scope and evaluation, while award notices can identify suppliers. Use the exact notice and dates. Do not assume one public contract represents a provider's wider business.
Check accountability evidence separately
Digital buying credentials need a narrower test than broad agency positioning. IAB UK's Gold Standard addresses named areas such as ad fraud, brand safety, supply-chain transparency and user experience. Participation can be recorded as one dated fact, but it should not become a blanket quality mark for every team, channel or campaign.
Ask each competitor which entity holds any certification, when its status was checked and how the relevant controls appear in day-to-day delivery. Retain the supporting page or certificate with the research sheet. If a badge cannot be traced to a current issuer record, leave it out of the comparison rather than repeating the agency's artwork or wording.
Compare evidence, not adjectives
- Use the same fields and blank values for missing information.
- Distinguish self-description, public record and independent research.
- Avoid converting absence of evidence into a zero score.
- Give every named agency its own source link.
End with unanswered questions for outreach: team availability, trading route, conflicts, account ownership, reporting and exit. Refresh volatile fields before publication. The research can inform positioning and a buyer shortlist, but it cannot support rankings or a claim that one competitor is best.
Add a media-buying capability layer
Generic agency labels hide material differences. Record which media the firm says it plans or activates, whether it describes programmatic buying, how it presents measurement and which markets its named office serves. Link every entry to that agency's own current page, then seek contract-level confirmation during procurement.
Also note what public material cannot establish: staff assigned to an account, access to inventory, negotiated terms, conflicts, control of platform accounts and the treatment of unspent media funds. A blank cell is more honest than an inference. Date the sheet and nominate an owner, because personnel, propositions and corporate records can change between research and appointment.
Before you act
- Name the client type, channels, territory and service boundary.
- Set one research date for every competitor entry.
- Write exclusion rules before looking at familiar names.
- Save the precise service page, not just the homepage.
- Give every named agency its own source link.
- Date the sheet and nominate an owner.
Common questions
Can a Companies House filing prove an agency is any good at media?
No. A filing record confirms aspects of legal identity, but it does not prove media skill, solvency or client satisfaction. Use it only as one dated fact about the entity, and keep the supporting record with the research sheet.
What does the Meta Ad Library actually show about a competitor?
It can show visible creative activity for named advertisers, with particular retention rules for some UK and regulated categories. It does not expose the agency behind every placement, the target audience, the full spend or commercial performance.
How should I treat an agency's Gold Standard badge?
Record participation as one dated fact, not a blanket quality mark for every team, channel or campaign. Ask which entity holds the certification, when its status was checked and how the controls appear in day-to-day delivery. Leave out any badge you cannot trace to a current issuer record.
