Where do your research participants actually come from? Why clients should scrutinise the sample supply chain in 2026

Research procurement tends to start with a specification: who do we need, how many, and by when? An equally important question often goes unanswered: where will those people come from, and what will have happened before they reach the interview?

A completed participant list does not tell you. Even people who have passed the same screener may have followed very different recruitment journeys: a direct invitation from an existing community, a public advertisement or a referral from an external supplier. What matters is not the channel’s label, but whether sourcing, selection and verification can be explained.

What is a sample supply chain?

The sample supply chain is the sequence of sources, intermediaries, selection processes and checks through which people gain access to a study. It runs from the first invitation through routing, screening and verification to participation, payment and feedback to the suppliers involved.

In quantitative online research, that chain may include panel owners, aggregators, routers and exchanges. A router directs a person towards an available survey; an aggregator combines access from multiple sources. The agency selling a sample is not necessarily the organisation that originally recruited the person. Each handover can make the chain harder to understand if source information and checks are not carried forward.

Qualitative recruitment does not necessarily follow this model. It may combine an owned database or community, public calls, targeted outreach, referrals and local recruitment partners. Studies typically involve fewer people, but more specific requirements for individual experience and suitability. The shared procurement question is: who carried out each step, and who is accountable for confirming the participant?

Why transparency matters more in 2026

Aggregation and digital marketplaces can expand access to audiences. They also introduce more handovers between the participant, the original source and the client. Generative AI can make plausible screener responses easier to produce. Tight B2B specifications and demanding timelines add pressure to supply apparently suitable people quickly. Together, these conditions make an explainable recruitment chain more valuable than a generic quality promise.

On 29 August 2026, the Insights Association published its Sample Supply Transparency Framework. Its 95 questions comprise 27 universal questions, 14 for owned first-party panels, 7 for aggregators, routers, exchanges and non-panel suppliers, 30 for research firms and agencies, and 17 for brands and end clients. Topics include sourcing, routing, validation, incentives, privacy, quality controls, subcontractor oversight and AI-assisted responding.

That allocation is useful: transparency cannot simply be delegated to the last supplier in the chain. Agencies and clients also need to know what they require, what information they review and how they handle deviations. The framework structures a discussion; it is not a certificate that a particular recruitment project is free of error.

What the benchmarking figures do—and do not—tell us

In the Insights Association’s H1 2026 Global Data Quality Benchmarking Report, agencies globally reported overall pre- and in-survey removals of 28.5%, compared with 21.2% reported by suppliers. The report identifies B2B as the most challenging environment. Germany and France were newly added to the country benchmarking.

These figures benchmark survey and sample data quality. They are not a measured fraud rate for qualitative recruitment, nor a forecast of how many people in your next interview study will be unsuitable. Removals can have different quality-related causes. The figures do not establish that every removal was a fraud attempt, or that the different rates provide a direct performance ranking of suppliers.

For qualitative buyers, the useful lesson is therefore not a transferable percentage. It is a procurement question: are quality problems identified, documented and fed back at different points in the chain? A supplier’s account of its controls is only meaningful if the basis of its measures and the stage at which they were collected are clear.

A completed screener is not provenance

A screener records answers to study-specific selection questions. Initially, it tells you what somebody has reported. It does not explain how they were recruited, which other studies they have been exposed to, whether several sources are supplying the same person, or what checks an upstream partner has already performed.

It helps to separate three layers. Provenance describes the recruitment journey. Validation checks relevant claims and actual experience. Participation readiness establishes whether the confirmed person can contribute in the intended format. None replaces the others. A source label does not prove suitability; a good personal conversation does not, on its own, give you visibility of every supplier involved.

The ESOMAR/GRBN Guideline on Online Sample Quality addresses participant validation, duplicate and fraud prevention, engagement, exclusions and transparency around sources, blending and routing. It concerns online samples, rather than serving as a complete manual for qualitative interviewing. Its questions about explainable sources and clear responsibilities are nevertheless relevant.

Owned communities and external sources: no automatic winner

An owned community can support direct contact and knowledge of previous participation. That does not automatically make it unbiased or appropriate for every audience. Relying exclusively on a familiar database may miss important perspectives. Frequent participation is not an automatic reason to exclude somebody either: what matters is the research question, relevant study experience and agreed project-specific exclusions.

External sources can extend reach and open access to new or specialist audiences. Problems arise when their use is undisclosed, duplicate contacts go unchecked or nobody can explain who rechecked the selection criteria. The right buying rule is not ‘owned is good, external is bad’. It is: disclose the sources, justify the selection, agree the controls and assign responsibility.

A good source is not simply a familiar source. It is an explainable source with appropriate controls and clear accountability.

B2B puts the chain to the test

In B2B research, role, employer and relevant experience cannot be reduced to a job title. A public professional profile may help establish the plausibility of a role or employer; it does not automatically prove budget responsibility, involvement in a particular purchase or hands-on use of a system. Where several suppliers are involved, someone still needs explicit responsibility for checking those decisive criteria.

Useful validation distinguishes, for example, between working in procurement, contributing to supplier selection and making the purchasing decision. It involves study-specific follow-up, public professional sources where appropriate, and documentation of unresolved criteria. If a relevant criterion cannot be adequately verified, the person should not be confirmed. A fluent answer does not prove professional experience; a brief answer does not prove deception.

Participant experience and incentives belong in the quality discussion

The chain does not end with a tick beside ‘qualified’. Clear invitations, appropriate incentives, dependable scheduling and understandable information about duration, recording and payment affect who takes part and how they experience participation. Repeated screening through different intermediaries or unexpected requirements can put suitable people off.

Checks should be proportionate to project risk and respect privacy. The aim is not to collect as many identity documents as possible. Buyers should ask what data is genuinely necessary, who receives it and when it is deleted. The 2026 MRS webinar description, ‘Safeguarding Qualitative Research: Understanding and Mitigating Fraudulent Participation’, addresses fraudulent, professional and over-researched participants, AI-assisted responses, better screeners, identity and location verification, and in-field validation alongside participant experience, privacy and responsibility across the supply chain. An event description is not a prevalence study.

Unusual behaviour warrants appropriate follow-up, not an automatic fraud label. Before fieldwork, agree how moderators and the recruitment team will handle unresolved suitability and how learning will reach the responsible source. Transparency needs a process for responding to problems, not just an explanation at the start of the project.

A buyer’s checklist: nine questions before commissioning

Use these questions in the brief and proposal discussion. Ask for concrete explanations and project-specific agreements, not just an assurance of ‘high quality’.

• 1. Which channels will supply the participants? Which owned, external or blended sources will be used, and why are they appropriate for this audience?
• 2. Who else is involved? Which subcontractors, intermediaries or routers will be used, and how will changes of source or further subcontracting be disclosed?
• 3. How is provenance retained? What is recorded about the recruitment journey, and what can the client receive without unnecessary personal data?
• 4. Who verifies which criteria? What relies on self-report, what is checked in a personal follow-up, and when is additional evidence proportionate?
• 5. How are duplicates and fraud risks handled? What controls work across sources, how is AI-assisted responding considered, and how are false alarms reviewed?
• 6. Which prior participation and exclusions matter? How is study history considered without treating regular participants as automatically unsuitable?
• 7. What will participants experience? Who explains time commitments, consent, recording and payment; are incentives appropriate and privacy responsibilities clear across the chain?
• 8. What happens when criteria remain unresolved or someone cannot attend? Who confirms suitability, how are risks communicated early, and what reminders, confirmations or backups have been agreed?
• 9. How does quality feedback work? Which removals and problems are documented with meaningful reasons, how are suppliers reviewed, and what are the limits of reported metrics?

What this means at Insight Fox

The WMM team has recruited participants for qualitative research since 2004; Insight Fox emerged as its specialist brand in 2026. We use owned participant, community and database channels alongside multi-channel recruitment. Channels are selected for the audience and project, rather than treated as proof of quality in themselves.

Selection includes a project-specific screener, validation and personal telephone follow-up to recheck key criteria and assess motivation and articulation for the intended format. For B2B, publicly available professional sources are used where appropriate. Participants are not coached towards expected answers.

Reminders, confirmations and backups are agreed for the project; risks should be communicated early and incentives set at an appropriate level. These are practical working methods, not a guarantee against every unsuitable participant. B2B recruitment, UX participant recruitment and focus groups each call for different checks according to the audience and format. The service pages and project assessment linked below provide a starting point for discussing those requirements.

Conclusion: transparency is quality control, not bureaucracy

Clients do not need to audit every intermediary personally or work through all 95 framework questions for every interview study. They should understand how people reach their research, which claims have been checked and who makes decisions when something remains unclear.

The useful question is not only ‘Does this person fit?’ It is also ‘Can we explain why they were selected and what happened on their way to participation?’ An explainable sample supply chain does not make research error-free. It makes the foundations open to scrutiny—and problems easier to discuss before they become findings.