In this article · 7 parts
Ask potential customers about something that actually happened before asking what they think of your idea. A conversation about their last relevant experience can reveal the problem, current alternatives, buying process and constraints. It cannot prove that they will buy from you, but it can prevent you building an offer around an imagined need.
This approach is useful when you have an idea but little evidence, or when people keep saying it sounds good without taking the next step. You can conduct the conversations separately, around available breaks and appointments; there is no need for a formal research event.
The aim is to understand a decision. You are not trying to persuade the person to endorse your plan.
1Decide what you need to learn
Choose one uncertainty before arranging a conversation. Perhaps you do not know whether people experience the problem often enough, whether somebody else controls the budget, or whether your proposed delivery window fits how they buy.
Write a question that could change your next action: “Do local stallholders need new photographs often enough to consider paying for a small repeatable service?” is more useful than “Do people like my photography idea?”
Define who can answer. Someone who recently prepared products for sale may have relevant experience. A friend who has never sold anything may be supportive but unable to explain that buying situation.
The Irish Local Enterprise Office recommends speaking to the intended market and warns against excessive reliance on friends and family. You can start with people you know, but record why they qualify as potential buyers and widen the sample beyond your immediate circle.
Do not attach a claim of statistical representativeness to a handful of accessible conversations. You are collecting detailed clues and testing assumptions, not estimating national demand.
2Invite a conversation without disguising the purpose
Explain briefly what you are exploring, why their experience is relevant and how much time you are requesting. Make declining easy. If you intend to discuss an offer as well, say so rather than disguising a sales pitch as research.
You might say: “I am exploring a small service for people preparing stock for local markets. I am trying to understand how you currently arrange product photographs. Would you be comfortable talking through your most recent experience? I am asking for around fifteen minutes, and it is fine to say no.”
That is suggested wording, not a promise that every interview will fit fifteen minutes or that someone will agree. Offer another suitable time or written answers if that makes participation easier. Do not ask for sensitive business records when a general account will answer your question.
If contact is promotional, calling it research does not settle the rules. Check the audience and channel before approaching people:
- Ireland: the Data Protection Commission's consent FAQ explains electronic-marketing consent and a limited existing-customer exception.
- United States: the FTC's CAN-SPAM guide explains commercial-email requirements. It is not a complete guide to calls, texts or state privacy law.
- Canada: CRTC guidance on CASL addresses consent, sender identification and unsubscribe requirements, including conditions for relying on implied consent.
A publicly visible address is not universal permission to market. Avoid scraped lists and bulk approaches whose legal basis you have not established.
3A conversation you can actually use
Begin with context: “When did you last need to deal with this?” If there is no recent or memorable occasion, do not supply one for them. Ask whether it comes up at all and why it has not required action.
Then follow the event in order:
- “What triggered it?” Establish why the person acted then.
- “What did you do first?” Find the real starting point, which may be asking somebody they trust.
- “What options did you consider?” Include doing it themselves or postponing it.
- “What did you choose, and why?” Learn what mattered in the actual decision.
- “What was difficult or disappointing?” Let them identify the issue before you name your solution.
- “Who decided and who paid?” Discover whether the speaker controls a purchase.
- “What would happen if you left it as it was?” Examine the cost of doing nothing without exaggerating it.
Ask about money, timing and frequency when relevant, but allow the person not to answer. “Roughly what did that cost?” is different from demanding an invoice. Record whether an answer is approximate.
Follow useful details instead of reading every question mechanically. If a person says that access to the premises caused the problem, ask what happened. That detail may matter more than the next question on your list.
4Keep your solution out of the first half
Explaining the idea too early can make the conversation about whether the person approves of you. Give them room to describe a problem you did not anticipate, or to explain why the current arrangement works well.
Avoid questions with a preferred answer built in. “Wouldn't collection make this much easier?” invites agreement. “How did you get the items there?” tells you what actually happened. “Would you pay €50?” is a hypothetical response. It is not equivalent to accepting a real offer with defined terms.
You can explain the idea later and ask what is unclear, what would prevent using it and what they would need to know before deciding. Record these reactions separately from their account of the past. Both are useful, but they are different kinds of information.
Do not argue with an objection. If you find yourself explaining why their current choice is wrong, return to understanding how they made it.
5Illustrative example: the problem changes shape
Made up, to show the idea
Illustrative amounts in this article are in euros. They demonstrate the method, not typical local prices.
Marta works split shifts and is considering a service organising stock for small market traders. She assumes the main problem is untidy storage.
One trader describes missing items during early-morning loading. Their storage is reasonably organised, but different people pack the vehicle each week. A second describes stock getting mixed between online orders and market stock. A third says their existing system works and they would not pay for help.
Marta should not summarise this as “everyone needs stock organisation”. The conversations point to different situations. She records loading coordination, channel separation and no current problem as separate findings.
She decides to investigate the loading issue with other traders who share that situation. A possible offer may involve a labelled packing system and a handover list, rather than reorganising an entire storeroom. Before offering it, she still needs to establish whether a buyer values the result, can authorise payment and will pay enough for the work.
The third conversation remains valuable. It helps define who is outside the likely customer group and prevents her treating every trader as a prospect.
6Turn notes into decisions
After each conversation, use four headings: actual event, current solution, constraint, implication to investigate. Separate a direct quotation from your paraphrase, and do not invent precise words afterwards. Ask permission before recording audio; short written notes are often enough.
Keep only the personal information you need, explain your purpose and protect the notes. Do not turn research participants into a marketing list without addressing the relevant requirements. You can often analyse the pattern using anonymous labels rather than names.
Look for repeated details, but also for the conditions around them. Several people in one social circle may share circumstances that other buyers do not. A problem mentioned by one unusually large customer may not justify a service aimed at small operators.
Where accounts conflict, test whether you are mixing different buying occasions, budgets or decision-makers. Do not average away a distinction that could determine the offer.
7Know when to move beyond conversation
Continue interviewing when an important question remains unclear and another relevant person could help answer it. Change direction when the problem you expected is absent or less important than another one. Move towards a bounded offer when you can describe the buyer, occasion, current alternative and unresolved purchasing question clearly enough to test.
There is no magic interview count. More conversation is not automatically better if it keeps you from making a clear offer. Conversely, an enthusiastic conversation should not trigger a large investment.
Your next action can be small: prepare an invitation for one suitable person and write the single uncertainty you want to resolve. If you cannot arrange a conversation this week, improve the question or identify where the right people can be reached with permission.
Finish every interview review with “Because of this, I will investigate ___.” That turns listening into a change in what you do next, while keeping the limits of the evidence visible.
General information to help you think things through. Not regulated financial, legal or tax advice.