Audience research
Part of Audience research for a lead network that separates both sides of the marketplace
Defining a lead network audience with requesters and buyers kept apart
Define an England lead network audience with a decision-based sampling frame, separate requester and buyer groups, and explicit evidence limits.
A lead generation networks target audience is the smallest group whose real decision can validate or reject a matching proposition. Define the requester and paying buyer separately, then recruit people who have recently faced the relevant event. Industry, age or company size is useful only when it changes that decision.
This method was prepared on 5 September 2026 for research in England. No participants were recruited and it contains no audience or demand finding.
What to take away
- Define the requester and paying buyer separately, because mixing them makes later evidence impossible to interpret.
- Screen for a recent event and the real decision, not industry, age or company size alone.
- Recruit through more than one doorway and record each channel to expose empty cells.
- Keep research participation separate from marketing consent, since promotional surveys fall under direct-marketing rules.
- Keep the definition only if participants recall the event and their evidence changes a product choice.
Name the unit and the event
Start with a sentence containing an actor, situation and action. For example: "an owner of an England-based plumbing business deciding whether to pay to see a household request received this month". It is narrower than "small trades", but it can be screened and studied.
Requester vs buyer unit
Requester side
- Unit
- Person with a request
- Decide
- Comparing or one quote
- Risk
- Mixed intent
- Evidence
- Screening answer
Buyer side
- Unit
- Enterprise or branch
- Decide
- Who is authorised
- Risk
- Mixed authority
- Evidence
- Job context
For the requester side, specify whether the person is comparing suppliers, seeking one quotation or asking only for information. For the buyer side, decide whether the unit is an enterprise, a local branch, a sole trader or an employee authorised to buy.
ONS publishes a 2025 business activity, size and location dataset with enterprises and local units by industry and region. It can frame where eligible businesses might sit. It does not identify current need, buying authority or willingness to purchase an introduction.
Turn assumptions into screening rules
Create an eligibility table before contacting anyone:
Eligibility table fields
- Rolewho decided or influenced
- Recencywhen the event happened
- Geographyregion of request or buyer
- Experiencesuccess, refusal, abandonment, first use
- Exclusionwho cannot answer
| Field | Rule to write | Evidence to retain |
|---|---|---|
| Role | Who made or influenced the decision? | Screening answer and job context |
| Recency | When did the event happen? | Month or bounded period |
| Geography | Where was the request or buying unit? | Region, not unnecessary full address |
| Experience | Success, refusal, abandonment or first use? | Category recorded before interview |
| Exclusion | Who cannot answer the decision? | Reason code |
| Illustrative worked example | England plumbing business owner deciding in August 2026 whether to pay to see a household request received that month | Screening answer, month of request, region, first use, channel |
Do not collect protected or sensitive attributes by default. The ICO's data-minimisation guidance says personal information should be adequate, relevant and limited to what the stated purpose needs. If a characteristic matters to inclusion or accessibility, document why and have the collection reviewed.
Recruit through more than one doorway
Existing customers are convenient but exclude people who refused, left or never reached the form. CIPHE, the Federation of Master Builders and Checkatrade can reach England-based trades, but they may omit informal or new entrants. Open social invitations favour people who notice the post and choose to reply.
Record the recruitment channel for every participant and look for empty cells in the matrix.
An indicative planning target is eight to twelve completed interviews per main cell, with no fewer than four in a smaller cell. The number depends on how many cells the matrix contains and how soon the evidence repeats.
The GOV.UK Service Manual's participant-finding guidance recommends actual or likely users, attention to recruitment bias and inclusion of disabled participants. It was written for government services; adapt its operational advice instead of treating it as a commercial-market standard.
Keep the invitation honest
State who is conducting the study, the subject, time, incentive, recording, data use and withdrawal route. ICO guidance distinguishes genuine research from direct marketing. Put any optional marketing choice outside the research invitation and assess it separately.
Decide whether the audience definition worked
After fieldwork, report who was invited, who completed, recruitment sources, ineligible cases, refusals and groups not reached. Do not convert a purposive interview sample into an England percentage.
Keep the definition only if participants can recall the event and their evidence changes a product choice. Revise it when the screen admits people without the decision, or when one channel supplies nearly all views. Stop if the team cannot recruit transparently without using contact details for an unstated purpose.
Before you act
- Write an actor, situation and action sentence for the unit.
- Build an eligibility table before contacting anyone.
- Collect only personal data the stated purpose needs.
- Record the recruitment channel for every participant.
- State recording, data use and withdrawal route in the invitation.
- Report refusals, ineligible cases and groups not reached after fieldwork.
Common questions
How should I define the target audience for a lead network?
Start with the smallest group whose real decision can validate or reject a matching proposition. Define the requester and paying buyer separately, then recruit people who have recently faced the relevant event. Industry, age or company size matters only when it changes that decision.
What should I record when screening participants?
Use an eligibility table covering role, recency, geography, experience and exclusion. Retain the screening answer and job context, the month or bounded period, the region, the category recorded before interview and a reason code. Do not collect protected or sensitive attributes by default.
When should I revise or drop the audience definition?
Revise it when the screen admits people without the decision, or when one channel supplies nearly all views. Keep it only if participants recall the event and their evidence changes a product choice. Stop if the team cannot recruit transparently without using contact details for an unstated purpose.
