Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Templates

Research Brief Template

A marketing research brief template: the decision, research questions, what is already known, audience and sample, method, budget, timing and deliverables.
Templateintermediate10 min readUpdated

What this template is for

Use the brief for any study that will take more than a few days or cost more than a few hours of staff time: customer interviews, a survey, a concept or message test, a usability study, desk research for a market entry, or a mix of these. It works whether the research is run in-house or commissioned from an agency or freelance researcher. For an agency, the brief is what they respond to with a proposal; for an in-house study, it is what the sponsor signs before recruitment starts.

The brief has neighbours with different jobs. The research design guide explains how to move from a decision to answerable questions, how to choose between methods and how to trade sample, timing and budget; it lists the brief's sections without a full worked fill. The customer interviews tutorial carries a discussion guide for one method. The survey and questionnaire design guide covers building the instrument once the brief has fixed what it must measure. The brief comes first and points to all three. For research that repeats in waves, the tutorial on setting up a brand tracking survey shows where this brief sits among the screener, competitor set, pilot and baseline decisions.

Writing a brief also protects the study from its sponsor. A sponsor who has written down, before fieldwork, what finding would change the decision finds it harder to discount an inconvenient result afterwards. That is the most valuable sentence in the document, and the template gives it a place.

The template

One brief per study, two pages at most. Keep the field names, so briefs can be compared and the next study can start from this one.

FieldWhat to write
DecisionThe business decision the research must inform, who makes it, the deadline, and what finding would change it
Research questionsThe primary question and up to three secondary ones, ranked, each answerable with evidence
What is already knownExisting data, past research and the assumptions the study should test
Audience and sampleWho must be heard, who is excluded, how they are recruited, how many and the known gaps
MethodInterviews, survey, test, observation or desk research, and why this fits the questions
Survey or guide notesTopics the questionnaire or discussion guide must cover, and any it must avoid
Budget and timingBudget, fieldwork window, reporting date and the decision meeting it feeds
Consent and data handlingConsent approach, incentives, data protection, retention and any sensitive topics
Deliverables and ownerWhat is delivered, in what form, to whom, and who signs off the brief and the findings

Attach background material, previous reports and data extracts rather than summarising them at length in the brief.

How to fill each section

Decision

Write the decision as a sentence with a verb and a date, then name who makes it and what finding would change their mind. "Decide before the summer range review whether to change the pack or the recipe" is a decision; "understand our repeat purchase problem" is a topic, and a brief built on a topic produces a report rather than an answer. The research design guide explains why the decision also sets the standard of evidence: a reversible, low-cost decision can rest on a few interviews, an irreversible one justifies more. The one-way and two-way door framework is a quick way to classify the decision before setting how much evidence the brief asks for. A brief that leaves the date out is the first cause in the diagnosis for research taking too long, which lines up each stage of a late study against the day the decision falls due.

Research questions

Turn the decision into questions the study can answer with evidence, and rank them. Ask about behaviour that has already happened rather than predictions: "what did you buy instead, and why" can be answered; "would you buy it again" invites a guess. A jobs-to-be-done framing, asking what the customer was trying to get done when they bought, often produces better questions than asking about the product. The research design guide gives a table of weak and answerable questions for common decisions. One primary question answered well beats eight answered shallowly.

What is already known

List the evidence already available: analytics, sales and loyalty data, reviews, support tickets, previous studies and published secondary sources. Say what each settled and what it left open. The audience research guide lists the sources most businesses already hold and rarely read. This field narrows the primary research to what is genuinely unknown and names the assumptions the study should test rather than repeat.

Audience and sample

Describe who the study must hear from, using the Segmentation or the Buyer Persona the decision concerns, and who must be excluded. State the sample frame (the list or panel participants are drawn from), the recruitment method, the number of participants and the known gaps. For a survey, the Sample Size follows from the precision the decision needs; for interviews, from the point where new conversations stop adding themes. The research design guide explains how to write down the trade-off between sample, timing and budget so the reader knows what the findings can bear. For interviews, groups and usability sessions, the tutorial on recruiting research participants turns this field into a frame, quotas and a neutral screener.

Method

State the method family and why it fits: qualitative methods for why and how questions, quantitative for how many and how much, a controlled comparison for questions about cause. Many studies need more than one, in sequence, usually interviews before a survey. Write the method as a preference open to challenge, since the researcher executing the study may know a better route; the research design guide explains why the method is chosen last, not first. For conversion research on one website funnel, where analytics, recordings, polls and usability sessions run in a fixed order, the tutorial on running a conversion research sprint shows how a brief like this one scopes the question and hands the findings to a test backlog.

Survey or guide notes

List the topics the instrument must cover, the comparisons the analysis will need, and any topics to avoid or handle carefully. Do not write the questions in the brief; that is the researcher's job. The survey and questionnaire design guide recommends starting from the analysis: the figures and comparisons the report will contain, with every question feeding one of them.

Budget and timing

Give the budget, including incentives and panel or recruitment fees, the fieldwork window, the reporting date and the decision meeting the findings feed. Work backwards from that meeting: findings that arrive after the decision have no value however good they are. The research design guide explains how a close deadline usually sets the method.

Describe how participants will be told what the study is and how their data will be used, how consent is recorded, the incentive, where personal data is held and for how long, and any sensitive topics or participants who need extra care. The MRS and ESOMAR codes set the professional standard for research, including voluntary and informed participation and keeping research separate from selling. The privacy and consent guide covers consent and data handling on the measurement side, and the consent section of the survey design guide covers anonymity for respondents. The tutorial on participant information and consent prepares the information, consent statements and deletion dates this field summarises. This page is not legal advice; data protection duties depend on jurisdiction.

Deliverables and owner

Say what arrives and in what form: a findings document, a readout, anonymised verbatims, data tables. Describe findings, not a slide count. Name who signs off the brief, who answers the researcher's questions during fieldwork and who receives the findings. The research design guide notes that the brief and the findings together become the record of what was known and decided, which makes the next study better.

Filled example

Below, the brief is completed for an invented business. The figures, dates and names are illustrative.

Illustrative example — not a real result

Pippinlark Cider: why supermarket trial buyers do not buy again

FieldPippinlark's research brief
DecisionBy 15 May 2026, the commercial director decides what to propose at the June range review with the main supermarket customer: a change of pack (from four 500 ml bottles to four 440 ml cans), a change of recipe (less sweet), or neither. If lapsed buyers mainly cite sweetness, the recipe changes; if they mainly cite format or price per drink, the pack changes; if they cite neither, both stay and the promotion plan is reviewed instead
Research questionsPrimary: why do shoppers who bought Pippinlark once in the last six months not buy it again? Secondary: what did they buy instead on their next cider purchase; what did they expect from the first purchase and how did the product compare; what part did the half-price promotion play in the first purchase?
What is already knownRetailer loyalty card data (via the retailer's data service): about 61,000 households bought Pippinlark for the first time between October and March, and 23% have bought again, mostly during promotions. Online reviews (n = 140) mention sweetness in about a third of comments. Assumptions to test: that the repeat problem is about taste; that promotion-led buyers were never in the brand's target segment
Audience and sampleAdults aged 18 or over who bought Pippinlark once in the last six months and not since. Exclude people working in the drinks trade. Interviews: 12 lapsed buyers, recruited through the retailer's research panel with purchase verified, split between promotion and full-price first purchases. Survey: 400 lapsed buyers and 200 repeat buyers from the same panel for comparison. Gap: buyers at other retailers are not covered
MethodInterviews first (45 minutes, video call) to find the reasons and the language; then a short online survey to size how common each reason is and compare lapsed with repeat buyers. Open to challenge from the agency
Survey or guide notesCover the first purchase occasion, what else was in the basket, who drank it, the moment they decided not to buy again, what they buy instead and why, and reactions to the pack and to sweetness. Avoid presenting the two proposed changes as options; ask about experience first. No questions on drinking quantities beyond what the decision needs
Budget and timing£14,000 including incentives and panel fees. Interviews 7–17 April; survey in field 22–29 April; findings readout 8 May; decision meeting 15 May
Consent and data handlingParticipants told the study is for Pippinlark and is research, not a sales contact. Age confirmed at recruitment. Interview incentive £40; survey panel points at the panel's standard rate. Recordings held by the agency, transcripts anonymised before sharing, recordings deleted 60 days after the readout
Deliverables and ownerA findings document of no more than eight pages answering each question in order; a one-hour readout to the commercial director and marketing manager; anonymised verbatims tagged by reason; survey tables. Brief owner and day-to-day contact: marketing manager. Sign-off: commercial director

Common mistakes

  • A topic instead of a decision. "Understand our customers" produces a report nobody acts on. Name the decision, its owner and its date.
  • No finding that would change the decision. If nothing could change the sponsor's mind, the study is decoration. Write the deciding findings before fieldwork.
  • Asking people to predict. Questions about what people would do produce optimistic guesses. Ask about what they did.
  • Too many equal questions. Eight questions of equal weight get eight shallow answers. Rank them and protect the primary one.
  • Skipping what is already known. Primary research that rediscovers what the sales data showed wastes the budget. Review existing evidence first.
  • Method chosen first. A brief that begins "we want a survey" has closed off better routes to the answer. Choose the method from the questions.
  • Unstated sample gaps. A sample drawn from one retailer or from existing customers is a legitimate design, but only with the gap written down.
  • Findings after the decision. Research that arrives late has no value. Work the timetable backwards from the decision meeting.

The campaign brief that does the same job for a campaign, and the other planning documents, are in the templates library.

Terms used on this page

Frameworks

Where this page is used

Also related

Sources

  1. Market Research Society — MRS
  2. ESOMAR — ESOMAR

Related