Start with a segment that combines a consequential problem, an accessible buyer, a deliverable offer and sufficient evidence for the next commitment. Do not automatically select the largest sector or the group containing the easiest person to interview. The research should reduce uncertainty around a choice, not prove that every sector is attractive.
A request for “market research” without defined alternatives can produce a broad report that changes nothing. A better brief says: we have this offer and this delivery capacity, and we are choosing between two initial segments. We need to know what favours each one and what would justify delaying the choice entirely.
Define segments by behaviour, not industry name alone
“Small businesses” is too broad to guide recruitment or sales. Add the relevant workflow, complexity, purchase trigger and approval owner. Two organizations can share an industry and size while one buys locally when a problem arises and the other is bound to a centrally controlled contract.
Before commissioning Market & Offer Research, create a short segment card: who experiences the problem, who searches for alternatives, who approves spending, when the need arises and how it is handled today. Label unverified statements as assumptions. Repetition in a presentation should not turn a team belief into a fact.
Agree what the decision actually authorizes. Selecting a segment for a limited learning exercise is different from committing a full sales team or redesigning the service. The evidence threshold should fit that commitment rather than being described vaguely as “enough confidence.”
Separate eligibility gates from preferences
Some conditions cannot be offset by a high score elsewhere: your ability to deliver, legitimate access to buyers, or an available owner who can approve implementation. A large segment should not rank first when it requires operating capabilities the business does not have.
Once those gates are met, compare problem recurrence, consequence, existing alternatives, purchase complexity and buyer access. Record evidence strength beside each judgment. A high attractiveness score based on speculation is not equivalent to a moderate score supported by multiple observations. A single total can hide that distinction.
Competitive Analysis helps describe the alternatives a segment faces. Competitor presence or absence is not decisive by itself. Absence could indicate weak demand, difficult delivery or an opportunity requiring further investigation.
Worked hypothetical example: building-equipment maintenance
A fictional company is considering non-medical building-equipment maintenance for independent clinics or small hotels. This is not a claim about either real sector, and medical-device maintenance is outside the example. The team can cover a limited geographic area but cannot promise round-the-clock emergency response.
| Research dimension | Independent clinics | Small hotels |
|---|---|---|
| Assumed need | Reduce disruption to reception and workplace facilities | Coordinate rooms and shared-area maintenance |
| Buyer to verify | Owner or operations manager | Manager or facilities lead |
| Eligibility question | Is the requirement limited to non-medical equipment? | Can required response hours be supported? |
| Alternative to investigate | Call-out technician or existing contract | Internal staff or external provider |
| Missing evidence | Contract review timing and approval authority | Effect of room occupancy on scheduling |
This explains why counting organizations is insufficient. Hotels may appear attractive in theory but be unsuitable for the current response offer. Clinics may be easier to reach while their actual requirements fall outside scope. Narrowing a segment before selecting it can therefore be a valid result.
Suppose early research identifies centralized purchasing within some of the proposed clinic group. The conclusion is not “clinics do not buy.” Redefine the candidate segment around independent purchasing authority, then investigate whether enough relevant organizations can be reached for a bounded test. That is a refinement of the audience definition, not an attempt to conceal disappointing evidence.
The eventual decision should identify excluded customers as clearly as included ones. For this example, the initial offer could explicitly exclude emergency response and specialist medical equipment. That allows sales and delivery teams to recognize unsuitable enquiries rather than discovering the mismatch after a proposal is sent.
Choose evidence that fits each unknown
Original public sources can establish published definitions and counts if needed, subject to checking year and unit. Interviews can explore users’ circumstances and context. GOV.UK interview guidance. Your sales history describes the customers you reached, not automatically the entire market.
Collect contradictory evidence too. Record people who decline, do not have the problem or already have an acceptable alternative. Do not remove them because they reduce enthusiasm. Document recruitment channels: drawing all participants from the founder’s contacts can make future access appear easier than it will be.
Confirm sector access and an accountable research owner before promising a study schedule. If recruitment is uncertain, make feasibility a first-stage output. A polished final report cannot compensate for interviewing people who do not match the decision being investigated.
Request a conditional recommendation
The report should lead to a clear outcome: test a narrow segment, obtain specific missing evidence, modify the offer, or stop investigating an option that does not fit current capability. State what would change the recommendation and who has authority to accept it.
Once the audience and offer are defined, Go-to-Market Strategy can address access channels and sales readiness. Avoid substituting “which advertising platform?” for “whom should we serve first?” before understanding the buyer and purchase process.
Acceptance criteria:
- Segment definitions make inclusion and exclusion clear.
- Operational gates are separate from attractiveness scores.
- Judgments have dated sources and explained confidence.
- Sample, access and data limitations remain beside the conclusions.
- Market counts and revenue projections are not invented.
- The recommendation names a decision or experiment and an owner.
Frequently asked questions
Should we choose the largest sector?
Not necessarily. You may be unable to reach its buyers or deliver its requirements. Size is one input; operating and commercial fit must also be established.
Can a few interviews support a decision?
They can uncover important unknowns or challenge an assumption, but do not establish market prevalence. Match the strength of the commitment to the evidence and its limits.
What if the two segments appear tied?
Identify the missing evidence most likely to change the choice. A small access test or an operational feasibility check may be more useful than adding more scoring categories.
Can the answer be not to proceed?
Yes. If both segments require unsupported capabilities or essential gates remain unresolved, postponing or revising the offer is a legitimate research outcome.
Provide your offer, present capacity and actual alternatives when requesting the study. That makes market research a decision process rather than a general description of an industry.
Method reference
- GOV.UK: Using in-depth interviews. Real sector counts require separate original sources; none are claimed in the example.
