Audience research
Part of Researching buyers, operators and affected people separately for privacy first measurement
Defining a measurement research audience by decision role and recent behaviour
Define a privacy first measurement target audience in England through decision roles, recent behaviour, exclusions and an auditable recruitment frame.
A privacy first measurement target audience is not "UK marketers" or "people who value privacy". It is a set of people selected because their recent responsibilities and experiences can answer a defined decision. For an England launch, that usually means separating buyers, measurement operators, assurance reviewers and people affected by data collection.
No interviews or surveys were carried out for this article. The process below is a recruitment method built from records reviewed on 5 September 2026, not a claim about audience size or demand.
What to take away
- Define the decision first, then recruit people whose recent responsibilities can answer it.
- Use observable behaviour and recent experience, not labels like privacy-minded or UK marketer.
- Split the sample by decision rolebuyer, analyst, reviewer and affected user each provide different evidence.
- For England, clarify whether eligibility depends on workplace, operating site or service location.
- Minimise recruitment data and separate contact details from session notes before collecting anything.
1. Write the decision and unit of study
Begin with the choice the research must change. A supplier testing its onboarding may study individual analysts. A team testing procurement barriers may need organisations as its unit, then recruit several roles inside each one. Mixing those units produces misleading counts.
Choose the unit of study
What decision must this research change?
individual analysts as unit
organisations, then several roles
The Government Digital Service (GDS) research planning guide says research planning should turn assumptions into questions and identify the groups required to answer them.
2. Build criteria from behaviour, not flattering labels
Use observable criteria: responsibility for an English website or app, involvement in a measurement change during a stated period, participation in a consent review, or authority over a relevant budget. Record disqualifiers too. Someone who follows analytics news but has not worked on a measurement decision may not answer a procurement question.
Observable criteria vs disqualifiers
Include
- Responsibility
- English site or app
- Measurement change
- In stated period
- Consent review
- Participated
- Budget
- Authority over it
Disqualify
- Responsibility
- No live property
- Measurement change
- News interest only
- Consent review
- Never involved
- Budget
- Interest without authority
GDS guidance on finding research participants recommends actual or likely users, clear recruitment criteria, a spread that includes access needs, and warns against recruitment bias.
3. Split the sample by decision role
Use a blank frame before deciding numbers:
Sample split by decision role
Role
- Commercial owner
- Buying trigger
- Analyst or engineer
- Data gaps, effort
- Privacy or legal
- Proof for review
- Affected service user
- Comprehension, effect
Evidence sought
- Commercial owner
- Interest, no authority
- Analyst or engineer
- Tool familiarity alone
- Privacy or legal
- Policy knowledge only
- Affected service user
- Staff speaking for users
Key exclusion
- Commercial owner
- Analyst or engineer
- Privacy or legal
- Affected service user
Split the sample by decision role
| Role | Recent experience required | Evidence sought | Important exclusion |
|---|---|---|---|
| Commercial owner | Approved or declined a relevant project | buying trigger and alternative use of funds | interest without authority |
| Analyst or engineer | Implemented or reconciled measurement | data gaps and operating effort | tool familiarity alone |
| Privacy or legal reviewer | Assessed an actual data flow | proof required for review | general policy knowledge only |
| Affected service user | Encountered the collection or choice | comprehension and practical effect | staff speaking for users |
Recruiting one group cannot fill evidence gaps in another. A buyer can describe approval, not what every visitor understood. An analyst can explain an implementation, not issue a legal conclusion.
4. Add geography without overstating it
For an England question, define whether eligibility depends on the participant's workplace, the organisation's operating site or the location of the measured service. A UK head office can serve users elsewhere, so postal address alone is weak classification.
UK ICO guidance is relevant throughout England, while a result collected across all four nations is UK-wide unless the sample permits separate reporting. Global platform documentation may identify technical settings, but it cannot establish the views of English buyers.
5. Minimise recruitment data
Screen only for facts needed to place a participant in the frame. The ICO's data protection principles include purpose limitation, data minimisation and storage limitation. These are legal principles, not optional research etiquette.
Keep contact details apart from session notes where practical. Define access, retention and deletion before recruiting. If screening could reveal health, ethnicity or other sensitive information, involve the organisation's privacy lead and researcher before collection.
The result should be an auditable frame showing who qualifies, who is deliberately included, who is missing and which decision each group informs. Until people are recruited and evidence is collected, it remains a hypothesis about whom to study, not a validated audience description.
For example, the frame can have fields for participant ID, decision role, observable criterion met, disqualifier checked, geography basis, consent reference and contact status.
Before you act
- Write the decision the research must change.
- Choose the unit of studyindividual or organisation.
- Set observable criteria and record disqualifiers.
- Split the sample by decision role.
- Define geography for England eligibility.
- Minimise recruitment data and set retention.
Common questions
What criteria should I use to screen participants?
Set a time window for each criterion, such as "in the last 12 months", so recent behaviour is testable. Add disqualifiers for people who only follow the topic. Keep a written record of why each person was excluded.
How should I handle geography for an England launch?
Decide if you need England-only findings or UK-wide. If England-only, screen on the operating site or service location, not a head office postcode. If the sample cannot support separate reporting, label the result UK-wide.
