Audience research
Part of Audience research for a website project that keeps buyers and end users apart
Target audience for a website project: define buyers and end users before recruiting research participants
Define separate buyer and end-user populations for a website project in England, then recruit with clear consent and link the evidence to one decision.
Start with the decision the research must change. Then create one population for organisational buyers and another for the people who attempt the service task. Keep the two apart until the evidence says they overlap.
What to take away
- Name the decision and its owner before you recruit anyone.
- Define buyers by decision authority, not by invented personality traits.
- Define end users by an eligible task and a time window.
- Include people who used telephone, face-to-face or assisted routes after a digital failure.
- Tell participants what the research is for, how long data is kept and that they can withdraw.
1. Name the decision and owner
Write the next choice in operational terms: keep or remove a channel, rewrite a content group, prototype a transaction, or approve a web design brief. Name one owner who can act on the result. Research without an owner produces a report that nobody uses.
Add an expiry date. Audience facts go stale, so state when the definition must be checked again. A population defined for one launch should not be reused for the next without review.
2. Build the buyer frame
List the roles that set scope, budget, risk or acceptance: service ownership, procurement, content, accessibility and technical delivery. For each role, record the organisation type, the actual decision authority, the England location rule, recent involvement and any conflict of interest.
A role is a recruitment category, not a persona. Our worked method for buyer persona cards shows how to record authority and confidence without inventing character traits.
Worked example, with invented details: a buyer frame for an English local authority housing repairs portal. The service manager approves discovery, the procurement officer controls the contract, the repairs team leader owns the tasks, and a tenant panel representative speaks for users.
3. Define the end-user population
Describe an eligible task and a time window: someone who recently attempted a specific application, purchase or account action. Include people who used telephone, face-to-face or assisted routes after a digital failure, because they reveal barriers that completed online journeys hide.
Government guidance on learning about users and their needs says that, in discovery, you should find out who your likely users are and what they are trying to do. It adds that you should learn how they currently do it, including which services or channels they use, and the problems or frustrations they experience.
Check access needs here too. A population that excludes assistive-technology users will miss the accessibility problems that the web accessibility checklist is meant to catch.
4. Select recruitment routes deliberately
Choose more than one route where practical: an authorised invitation to existing users, a community organisation notice or a specialist supplier. Each route reaches different people, so record who each one cannot reach.
For every route, log the invitation count, refusals, eligibility failures, non-response, attendance and withdrawal. The table shows an illustrative log with invented counts.
Select recruitment routes deliberately
| Recruitment route | Invitations | Refusals | Eligibility failures | Withdrawal |
|---|---|---|---|---|
| Authorised invitation to existing users | 40 | 6 | 3 | 1 |
| Community organisation notice | 12 | 1 | 2 | 0 |
| Specialist supplier | 20 | 4 | 1 | 1 |
Do not reuse a marketing list merely because contact details exist. Invitations to research, participation information and later promotional contact have different purposes. The data duties on a website project explain how purpose and role shape what you may do.
5. Make participation workable
Tell people what they are agreeing to before the session. They must understand who is doing the research, why, what data you collect and what will happen. They must also know how results will be used, that taking part is voluntary and how long data is kept.
The GOV.UK Service Manual on informed consent recommends giving participants an information sheet during recruitment. Consent is needed from all participants, even those who work for your organisation. Say whether the session is observed and who is watching, and name the data controller.
Under UK GDPR you must have a valid lawful basis to handle personal information, and you must determine and document it before you start using it. The ICO lawful basis guide covers the options.
State the format, access adjustments, recording choice, incentive and expenses in the same document. An incentive does not buy a positive answer, so record it and offer alternative formats and timing.
6. Set the decision threshold
Define what evidence will trigger further investigation, a prototype change or a stop. An interview can reveal a credible problem, but it cannot show how common the problem is. Write that limit into the conclusion.
Publish the population definition, recruitment routes and missing groups with any finding. If the sample cannot reach a critical user, or the buyer owner cannot act, say so. Pair this with the method in customer interviews and the wider audience research guide.
Before you act
- Name the decision, owner, evidence needed and expiry date.
- Build the buyer frame with roles, authority, location and conflicts.
- Define the end-user population by eligible task and time window.
- Select several recruitment routes and record who each cannot reach.
- Prepare an information sheet covering purpose, data, retention and withdrawal.
- Set the decision threshold and publish the population definition with missing groups.
Common questions
What should I do before recruiting anyone for website research?
Name the decision the research must change and its owner. Then create one population for organisational buyers and another for end users. Keep them separate so that each group's evidence stays clear.
How do I define the end-user population for a website?
Describe an eligible task and a time window, such as a recent application, purchase or account action. Include people who used telephone, face-to-face or assisted routes after a digital failure.
What should I tell participants before they agree?
Tell them who is doing the research, its purpose, what data you collect and how long you keep it. Say that taking part is voluntary and that they can withdraw at any time.
