What experts know that buyers and reports do not
Most market questions can be started at a desk. The tutorial on running a desk research pass works through internal data and published sources and ends with a list of what no source answers. Some of those unknowns are about why customers choose, and they belong to the customer interviews tutorial. Others are about how the market itself works: who buys through whom, how a purchase is specified and approved, which kinds of buyer behave differently, and why the published figures look the way they do. Buyers rarely know these things, because each sees only their own purchase. Reports rarely say them, because they are not easily counted.
People who stand in the middle of a market see many purchases at once. A distributor knows which customers order through it and which buy direct. An installer knows what is specified, what is swapped at the last minute and why. A consultant who advises buyers knows how decisions are made and who signs them off. A trade journalist knows which companies are growing and which stories the industry tells itself. Their knowledge is especially useful for three jobs: drawing the structure of a market before sizing it, testing the assumptions in a bottom-up estimate, and explaining a gap between two published figures.
| Question | Best first source | What experts add |
|---|---|---|
| How large is the market? | Published figures and the business's own data | Whether the published definition matches how the trade sees the market; counts that registers miss |
| How does buying work? | Expert interviews | The route from need to purchase, who specifies and who approves, through which channels |
| Which segments exist? | Expert interviews, then customer data | The kinds of buyer that behave differently, as the trade sees them |
| Why do buyers choose one supplier? | Customer interviews | Context only; experts report what they believe buyers think |
| What will competitors do next? | Public sources only, as the competitive intelligence guide sets out | Nothing confidential; only what is public or general knowledge in the trade |
Who counts as an expert
An expert, for this purpose, is anyone whose position lets them see more of the market than a single buyer does. The useful types differ in what they see and in what they want from the conversation, and both shape how their views are weighed later.
- Channel and distribution. Wholesalers, resellers and distributors see order patterns across many buyers. They know the market as it flows through them and may have a commercial interest in how a new supplier enters it. The guide to channel partner marketing covers working with such partners as a route to market; an expert interview is a different thing, a research conversation that asks for knowledge rather than a partnership.
- Installers, contractors and service providers. People who fit, maintain or operate products in the field know what actually gets used and what fails.
- Consultants and advisers. People who help buyers choose know the decision process and the criteria used.
- Trade journalists and analysts. People who report on a sector see its companies, its events and its arguments over time.
- People who have worked in the industry. Former managers and specialists bring a long view of how the market works. They also bring obligations to former employers, which the lines below respect.
Write a short profile of the experts a study needs in the same way the guide to recruiting research participants describes participants: by what their position lets them see, not by job title alone. Aim for experts in different positions, so that each view can be checked against someone who sees the market from another side.
Finding them and introducing yourself honestly
Experts are found through the trade's own visible places: trade press bylines, conference speaker lists, association directories, professional networking profiles, the business's existing suppliers and customers who know the industry, and introductions from people the team already knows. Whatever the route, the introduction follows the principles the site's research guides draw from the MRS and ESOMAR codes, as the methods hub describes them: participation is voluntary and informed, people are told who is collecting the information and why, and research is never a disguise for selling.
In practice that means saying who you are and which business you work for, what the conversation is about in general terms, roughly how long it will take, how notes will be kept and used, and that the person is free to decline or to skip any question. It also means not posing as anything else. The competitive intelligence guide is clear that pretending to be a customer, journalist, student or job applicant to extract information someone would not share if they knew who was asking is misrepresentation. A research conversation that opens honestly is also a better conversation, because the expert can judge what is appropriate to say.
Keep the invitation free of any sales pitch. An expert who is also a potential customer or partner may well become one later, but a research conversation that turns into a pitch has stopped being research.
What to ask and what never to ask
Ask about structure, process and patterns. These are what experts see well and can share without breaching anyone's confidence. The five forces model is a useful frame for structure questions, as its page describes: how hard current competitors fight for buyers, how easily new suppliers enter, what substitutes buyers use, how much power buyers and intermediaries hold, and how much the suppliers the business depends on can dictate terms. Questions about kinds of buyer feed the business's SegmentationDividing a market or customer list into groups that need different offers or messages. and its Ideal Customer ProfileA description of the type of organisation most likely to buy, succeed with the product and stay, used to choose target accounts., and questions about volumes and replacement cycles feed a bottom-up market size.
| Ask | Never ask |
|---|---|
| How does a typical purchase happen, and who is involved in specifying and approving it? | What a named competitor is going to charge, or the margins it works on |
| Which kinds of buyer behave differently, and why? | A competitor's plans, tenders, bids or customer lists that are not public |
| Where do published figures for this market seem right or wrong to you, and why? | Confidential information from a former employer |
| What has changed in the market recently, and what is changing now? | Anything the expert has said they cannot share |
| Which substitutes do buyers use instead, including doing nothing? | Questions that assume the expert will break a confidence |
| What would make a buyer switch supplier, as you have seen it happen? | Personal details about third parties that the study does not need |
The right-hand column follows the competitive intelligence guide's lines exactly. No confidential information from insiders: someone who has worked for a competitor should not share its confidential material, and the conversation should never be designed to draw it out. No exchange of competitively sensitive information: competition authorities, including the UK's Competition and Markets Authority, publish guidance on the risks of competitors sharing future prices, margins or plans, whether directly or through a third party such as a trade body or a consultant. An expert interview must never become that third party. Published prices and what any customer can see are fair to discuss. This is general direction rather than legal advice; where a conversation or a topic is in doubt, ask legal counsel before it happens, and write the study's rules into the discussion guideThe plan for a research interview or group: a warm-up, the episode to reconstruct, open questions about past behaviour and probes, used as a plan, not a script. so that everyone running interviews knows them.
Running the conversation
Prepare a short discussion guide, built in the same way as the one in the customer interviews tutorial: an introduction that restates who you are and why you are asking, a warm-up about the expert's role and what they see, the core questions in order of importance, and a close that asks what you should have asked and whom else you should speak to. Expert conversations differ from customer interviews in one respect: you want the expert's general picture, not their personal story, so ask for patterns and then for an example that illustrates each one.
Ask for the basis of every claim. "How do you know?" and "How often have you seen that?" separate what the expert has observed from what they have heard or assume. Ask for ranges rather than single numbers, and ask what would make the number higher or lower. Note which statements are observation, which are hearsay and which are opinion.
Watch for the conversation drifting across the line. If an expert starts to describe a competitor's confidential prices, bids or plans, stop the topic politely, say that you cannot discuss it, do not write it down, and move on. If something has already been said, leave it out of the notes and the findings, and ask legal counsel if you are unsure what to do. Ending a call early is always an option.
Experts' names, contact details and notes about what they said are personal data. Keep only what the study needs, store notes under a code rather than a name where you can, and handle them as whoever owns data protection in the business directs. Tell the expert whether they will be named or quoted anywhere, and keep to what was agreed.
Weighing what experts say
An expert's view is evidence of what that person believes, from where they stand. It is not a measurement. The qualitative versus quantitative comparison makes the general point that conversations explain and measurements count; a handful of expert views can show how a market is structured, but they cannot say how common something is.
Record each view that bears on a figure as an assumption on the market sizing worksheet, with the expert's code as its source and a confidence rating that reflects the expert's position, the basis they gave and whether other experts agreed. Then reconcile it as the market sizing guide describes: compare routes that do not share assumptions, and read the gap as information about which assumption to check. These questions help set the confidence:
- Where does the expert stand? A distributor sees the buyers who use distributors, and may not see those who buy direct. Every position has a blind spot.
- What does the expert want? A trade body figure, as the market and competitor research hub notes, comes from people with an interest in their industry looking healthy, and individual experts have interests too.
- Does it agree with other evidence? A view that matches published figures, the business's own data and other experts in different positions carries weight. A view that stands alone becomes a question for the next round of research, not a planning assumption.
One expert's view recorded as a figure and never weighed this way is one of the causes the diagnosis for a challenged market size estimate traces when a board questions a size. The academy checkpoint on repositioning, trade evidence and rebrands tests weighing what experts and the trade say against customer evidence when a brand's positioning is in question.
Structure findings, such as a channel that controls most of the buyers, go into the situation analysis alongside the competitor analysis, and any assumption that ends up setting a KPIA metric tied directly to an objective. target should be marked so that it is checked first when results come in. For a business comparing several countries, the same expert conversations inform the demand, competition and cost-to-serve signals in the market entry and prioritisation guide.
A composting equipment maker sizes a new market (illustrative)
Gablovex Composting interviews the trade to size hospital kitchens
Gablovex Composting, an invented maker of on-site food-waste composting units, wants to know how many hospital kitchens could use one. Desk research gives a count of hospitals but nothing on how many run their own kitchens or how their food waste is handled. The team interviews 4 waste contractors, 1 facilities consultant and 1 former trade journalist, each introduced honestly as a research conversation for Gablovex with no sales pitch.
The contractors agree that most hospital catering is outsourced, which changes who decides: the caterer's site manager specifies equipment, and the hospital's facilities team approves it. The consultant estimates that around a third of sites have the floor space and drainage a unit needs; one contractor thinks it is nearer a quarter. The former journalist explains that a published figure on hospital food waste counts patient meals only, not staff restaurants, which is why it looked low.
Each view goes onto the market sizing worksheet as an assumption with its source code and a confidence: the outsourcing pattern is rated high because all 4 contractors describe it, and the floor-space share is rated low and carried as a range of 25% to 33%. During one call, a contractor begins describing the price a rival submitted in a recent tender. The interviewer stops the topic, says it is not something Gablovex can discuss, leaves it out of the notes and ends the call a little early, with thanks. The sizing reports a range and names the floor-space share as the assumption a site survey should narrow next.
Common mistakes
- Treating one expert as the market. A confident view from one position, with its blind spots, entered into a plan as fact.
- Asking experts what customers think. Experts can describe patterns; buyers' reasons come from buyers.
- A vague introduction. Not saying who is asking and why, which is misrepresentation and produces guarded answers.
- Asking for a competitor's prices or plans. Crossing the line the competitive intelligence guide draws, including through an expert acting as a go-between.
- Writing down what should not have been said. Confidential information that slips out belongs nowhere in the notes or findings.
- No basis recorded. Observation, hearsay and opinion mixed in the notes, so nobody can later tell which views to trust.
- Views never weighed. Expert opinions that skip the worksheet and the reconciliation and go straight into a number.
