The Silence Between the Lines: A UK Buyer's Method for Reading Research Proposals

A research proposal is rarely just a price. It is a description of the decisions a supplier intends to make on your behalf: whom to approach, how hard to try, how to correct the sample, who writes the questions and what will be reported when the data are thin. Yet many UK buyers compare proposals as though they were quotations for a commodity, focusing on the headline fee, the total sample size and the delivery date. The gaps in a proposal then become visible only after fieldwork has begun, when the budget is committed and the timetable is tight.

Reading for omissions is not a hostile exercise. It is a procurement discipline. A proposal that is silent on weighting, incentives or refusal handling is not necessarily weak, but it is asking the buyer to accept a default. The purpose of a structured read is to make those defaults explicit before they become contractual.

Why the silences matter

A credible supplier should be able to explain the trade-offs behind its design. If a proposal promises 1,000 online interviews among UK adults for a modest sum, the reader is entitled to ask what that figure buys in terms of source, tenure and subgroup coverage. If it promises a representative sample, the reader should ask representative of what, according to which data, and with what tolerances.

The silence in a proposal is evidence of where risk is being transferred. That transfer may be perfectly reasonable — no buyer can specify every operational detail in advance — but it should be a conscious decision rather than an accidental one. The commissioning email is often the cheapest place to improve evidence quality, because a well-framed clarification costs nothing compared with re-running fieldwork.

A reading method for tender documents

Begin with the commercial summary, but do not stop there. Read the method narrative twice: once for what it claims, and once for what it avoids. Then read the annexes, assumptions and exclusions, where operational detail is often buried. Keep a simple two-column note: what the proposal states, and what it leaves to implication. The second column is the agenda for clarification.

Pay particular attention to five areas where omissions are common and consequential.

Weighting and the population being described

A proposal may say the sample will be weighted to be representative. It should also say which variables will be used, where the target distributions come from, whether the weights will be capped, and how the supplier will report design effects. If none of this appears, the buyer should assume that weighting will be decided later, possibly by habit. For board-level decisions, an unweighted or lightly weighted sample can shift the story more than the margin of error suggests.

Incentives and the shape of the response

Incentives influence who takes part, not merely how many. A proposal that offers a prize draw may attract a different profile from one that pays a modest cash incentive or makes a charity donation. Ask whether the incentive is disclosed, tested, and monitored for its effect on response composition. Silence here is not neutral; it means the supplier has a default that the buyer has not seen.

Refusal handling and non-response

Response rates are not a virtue in themselves, but refusal handling is a design choice. Does the proposal describe call attempts, re-issue rules, replacement sampling, soft refusals, or the treatment of partial completes? If not, the buyer cannot tell whether the final sample will be shaped by persistence or by convenience. In UK business surveys especially, a proposal that ignores refusal handling is likely to over-represent the easily reached.

Sub-group sample sizes

The total sample size is the least interesting number in most proposals. Boards rarely ask only about the national picture; they ask about region, sector, size band or customer type. A proposal should state expected base sizes for the cuts that matter, together with the minimum reporting threshold and the confidence intervals those bases imply. If the proposal promises analysis by subgroup but gives no base sizes, the buyer should ask for them before signing, not after the tables arrive.

Who actually writes the questionnaire

Questionnaire authorship is often invisible in proposals. Is the instrument drafted by a research methodologist, a sector specialist, an account manager or a client team? Will it be cognitively tested or piloted? Are questions drawn from a validated source, and if so, will the wording be preserved? These questions matter because small changes in wording can alter the story. A proposal that treats questionnaire design as an administrative step is a proposal that has not shown its method.

The question set to send back

Before signing, send a short clarification to every shortlisted supplier using the same wording. Ask for concise written answers, not a revised sales document. A workable set is:

1. Please set out the proposed weighting scheme, including variables, targets, sources and any capping.
2. What incentive will be offered, when, and how will its effect on sample composition be monitored?
3. How will refusals, non-contacts and partial completes be handled and reported?
4. What are the expected unweighted base sizes for the five most important subgroup cuts?
5. Who will draft the questionnaire, who will sign it off, and what testing will take place?
6. Which questions, if any, are taken from existing surveys, and will wording be preserved?
7. What will the supplier do if a subgroup falls below the agreed reporting threshold?
8. Which parts of the method are fixed in the contract, and which remain at the supplier's discretion?

The answers should be scored alongside price and sample size. A supplier that gives specific, bounded replies is usually easier to manage than one that offers fluent reassurance. Where answers reveal a gap, the buyer can decide whether to accept the risk, adjust the brief or reallocate budget.

Procurement as an insight discipline

The cheapest route to better evidence is often not a larger sample or a more elaborate analysis plan. It is a better commissioning conversation. Proposals should be read as arguments about method, not merely as offers. The buyer who notices the missing paragraph — on weighting, refusal handling or questionnaire authorship — is not being pedantic. They are protecting the decision that will eventually sit on the board agenda, and making sure the evidence behind it is described before it is produced.