Researching who reads a creator media kit in England without turning it into a sales call

Audience research for a creator media kit starts with a recent decision, not a demographic profile. Recruit an English buyer who recently requested or checked a commercial pack, a creator who maintained one, or an authorised representative who handled an enquiry.

Ask what happened, inspect records only with permission, and report findings for that sampled task. Do not turn a handful of conversations into an England-wide claim.

This guide has a narrow subject. A media kit is a creator-controlled commercial information pack, and a portfolio tool creates, hosts, updates or exports it. Research into web hosting, design software, influencer campaigns, UGC marketplaces or agencies may expose adjacent work, but it does not describe this niche unless the media-kit task can be isolated.

What to take away

State the decision before choosing a method

A project needs one decision that research can change. Examples include whether to alter an evidence panel, offer a downloadable rights schedule, or replace an enquiry form. "Understand creators" is too broad. A workable question is: can a buyer verify the owner, date, definition and limitation of a claim in the current pack without asking for an undocumented follow-up?

The GOV.UK Service Manual advises teams to turn assumptions into research questions and choose activities capable of answering them in its guide to planning user research. That guidance was written for government services. Here it supports a research discipline, not a claim that a private portfolio service meets a government standard.

Write the decision record before recruitment:

A filled-in example, using a fictional creator and buyer:

This example is illustrative, not a real case. Leave participant counts and thresholds blank until a research lead justifies them. The work should stop if the required population cannot be recruited without guessing identity, using a marketing list for an unstated purpose or exposing confidential creator material.

Separate the people around the pack

"Audience" hides groups: the creator supplies or approves information, and a representative acts only within recorded authority.

A brand buyer examines evidence for a commission, while an agency user may prepare a shortlist but is not automatically the budget owner. A portfolio-tool operator hosts or processes records, and a featured person, photographer, musician or rights holder can be affected without being a user. An audience member sees later published content, not necessarily the kit itself.

Recruit on observable experience in each role: for a buyer, eligibility might require requesting or reviewing a creator pack for an English business unit. For a creator, it might require a dated update, export or response to a commercial enquiry.

A social profile, follower total or self-applied label is not enough, as it cannot prove authority, location, age or buying power. Nor accessibility need or permission to contact.

GOV.UK guidance on finding user-research participants recommends explicit criteria, relevant experience and inclusion of disabled participants. Apply the useful method while recording that the page addresses public-service research. Recruitment should include people who use different formats or assistive technologies when those experiences bear on the decision, rather than asking one participant to speak for a group.

Construct a traceable sampling frame

A sampling frame is the list or route for approaching eligible people. Build separate frames for creators, representatives and buyers. A customer-support log might identify buyers who asked about an evidence field; a permissioned creator register might identify people who exported a kit.

Contract or invoice records may show which English unit made a purchase, subject to the agreed research purpose and access controls.

Companies House's company information service can verify narrow filed facts about an incorporated buyer or supplier. It cannot show who used a media kit, who holds procurement authority, or whether a current service exists.

The ONS UK business activity, size and location dataset for 2025 helps describe registered-business strata. Its industrial categories do not isolate creator media-kit decisions, and neither source supplies a participant list for this study.

Log each frame's owner, snapshot date, geography, coverage, known omissions and lawful access route. Then count sampled records, eligible records, invitations, responses, completed sessions and usable cases separately. A high response among a convenient group does not repair missing parts of the population.

An illustrative sampling-frame log:

This log is illustrative, not a real dataset.

Keep participation separate from commercial contact

An invitation should identify the researcher, purpose, activity, expected use of the material, incentive, recording plan and contact route. Participation must not be bundled with a creator pitch, buyer lead or permission to publish a portfolio. Agreement to a research interview is not permission for later sales messages, a testimonial, campaign contact or reuse of an asset.

Sample recruitment invitation:

Subject: Research invitation about media-kit checks

Hello [name],

I am [researcher name] at [organisation]. We are studying how buyers check claims in creator media kits. The session takes about 30 minutes online. We will ask you to walk through a recent pack and say what you checked. You will receive [incentive, if any]. We will record audio only with your separate consent.

Reply to [research contact] to ask questions or decline. This invitation is for research only, and we will not add you to a marketing list.

Sample consent wording:

I have read the information sheet. I understand the purpose is research. I agree to take part and to the recording described. I know I can withdraw before [date] and ask for my data to be deleted. I understand this consent is separate from any commercial or marketing contact.

The GOV.UK page on informed consent for user research covers information sheets, understandable formats, recording disclosure and withdrawal records. Its consent process is an ethical research reference.

A named privacy specialist must still determine the organisation's actual UK GDPR basis, roles, transparency and rights handling. Record legal consent for personal-data processing and the participant's informed agreement to take part separately.

The ICO explains how to identify direct marketing. A message framed as research can still be promotional if its purpose includes encouraging a commercial response. The research owner and a qualified privacy and PECR reviewer should classify the intended invitation route before names or contact details are loaded.

Design sessions around evidence, not opinions alone

Use neutral prompts about the last completed task. Ask the participant to reconstruct how a pack arrived, what they needed to decide, which fields they trusted, what they checked elsewhere and what happened when evidence was missing. Do not ask whether a proposed product sounds useful and report the polite answer as demand.

Session script with neutral prompts:

Show a fictional media kit if observation is needed. It can contain an expired rights line, an undefined audience figure, an ambiguous representative and a difficult-to-read download. The participant can explain what they would inspect or reject without exposing a real creator or live campaign. Any real document should be opened only under recorded permission, confidentiality and redaction rules.

Objective commercial claims also have an external boundary. CAP Code section 3 covers documentary substantiation for claims open to objective proof in marketing communications.

Whether a pack is an advert needs qualified assessment. The research exercise should still ask if a claim has a source, owner, definition, date and limitation. A participant's confidence is not proof that the statement is true.

Rights evidence requires its own prompt. GOV.UK notes that copyright can be licensed for specified uses. Ask how a buyer or creator identifies the work, licensor, channels, territory, term, edits and withdrawal route. Do not ask participants to upload third-party images, music or confidential contracts to an unapproved research system.

Make interviews and surveys accessible by design

Offer the invitation, information sheet, consent record, interview task and survey in formats participants can use. Ask about access arrangements without requiring a diagnosis. Test keyboard order, focus, labels, error messages, contrast, zoom and downloadable documents with appropriate expertise and representative users.

The Cabinet Office guide to accessible communication formats recommends planning alternatives and using clear, usable formats. It is government guidance rather than a universal private-sector specification. The research team should combine it with a named accessibility review and testing of the actual instrument.

Remote work creates security decisions as well. Invitations, recordings, transcripts, screen captures and participant contact details should have access owners, retention rules and a tested deletion route. The NCSC's small organisation cyber-security guidance offers UK baseline practices, but it does not assess a chosen research supplier or configured account.

Choose a survey route on evidence

An online survey is useful when the sampling frame and question definitions are already credible. It cannot discover who the population is merely by circulating an open link. Before naming a tool, document respondent access, collaborator permissions, location and role of processors, export format, correction and deletion operations, accessibility evidence, support, contract and exit.

First-party pages show different, bounded capabilities.

These are supplier statements, not independent proof of suitability, accessibility or lawful processing for a media-kit study.

Use the same fictional questionnaire and acceptance tests for each candidate. Preserve unknowns about the buyer's exact account, edition, terms, support and configured data flow. A manual, buyer-controlled route remains a valid baseline when it better fits a small, permissioned frame.

Analyse without creating fictional personas

Separate observations, interpretations and decisions. The GOV.UK method for analysing a research session recommends recording what was seen or heard before turning observations into findings. Keep the participant identifier, task, artefact version and analyst note attached so another reviewer can trace the conclusion.

Use an evidence card instead of a persona. A filled-in example for a fictional buyer:

Small qualitative samples explain mechanisms and language in the study, but do not estimate prevalence.

Survey reporting needs the sampled, eligible, invited, started, completed and item-response denominators. Publish question wording, field dates, recruitment route, exclusions, duplicate handling, missingness and any weighting method. A named statistician should review claims intended to describe a wider population.

Turn findings into reversible changes

End with a decision ledger, not a stack of quotes. For each proposed change, cite the observation, affected population, evidence strength, counter-evidence, risk gate, owner and recheck date. A change to claims, disclosure, rights, privacy, safeguarding, accessibility or security stays on hold until its specialist has reviewed the actual wording and configuration.

The smallest useful outcome may be a clearer evidence field or a better export rather than a new tool. Test the amended media kit with fictional content first. If the buyer still cannot trace a claim, a creator cannot correct a rights record, or a participant cannot use the format, revert the change and investigate the failure.

Publication remains on hold. Before release, assign the writer and fact-checker, name the market-research, privacy, PECR, accessibility, safeguarding, advertising, consumer, intellectual-property, security and statistical reviewers, reopen every source and confirm the evidence map. No research finding in this draft has yet been produced with real participants.

Before you act

Common questions

What should a research project decide before choosing a method?

A project needs one decision that research can change, such as altering an evidence panel or replacing an enquiry form. The decision record must include the buyer decision, England rule, recent task, evidence, counter-evidence, owner and expiry. Leave participant counts blank until justified.

How can researchers keep participation separate from commercial contact?

An invitation should identify the researcher, purpose, activity, expected use, incentive, recording plan and contact route. Participation must not be bundled with a creator pitch or buyer lead. Agreement to a research interview is not permission for later sales messages or campaign contact.

What should researchers ask participants to focus on during sessions?

Use neutral prompts about the last completed task. Ask participants to reconstruct how a pack arrived, what they needed to decide, which fields they trusted, what they checked elsewhere and what happened when evidence was missing. Do not ask whether a proposed product sounds useful.