A team can describe its service in detail and still fail to understand why customers choose another supplier. Sometimes the reason is information missing from the website: working terms, evidence of experience or a clear explanation of how pricing is calculated.
An interview helps reconstruct the selection process through a person's actual experience. Its output informs website structure and advertising hypotheses; it does not confirm that the proposed idea will necessarily sell.
Do not try to learn everything about the audience in a single conversation. Define a few research questions:
For a new website, this helps determine page content. For advertising, it helps identify when the offer is relevant and which arguments deserve testing.
Interviews complement competitor analysis before launch: a competitor's website shows its offer, while a conversation with a buyer reveals how they perceived it and what they compared it with.
Select participants based on relevant experience: a recent purchase of a similar service, an active search for a supplier or a decision not to proceed after discussions. If the business already operates, do not limit interviews to its most loyal customers.
Separate groups with different needs. In B2B, the person initiating the search, the specialist assessing the solution and the manager approving the budget may evaluate an offer against different criteria. Do not merge their answers into a single “customer profile”.
For international markets, record the country, communication language and purchase conditions. Differences may relate to local procurement procedures or required documents, not just the translation of a page.
There is no universal interview count for every business. Start with a small series in one segment, analyse the answers and refine the questions for the next series. Recurring observations help identify themes, but do not tell you what percentage of the market shares an opinion.
Instead of “Would you buy this service?”, ask “Tell me about the last time you chose a supplier.” Explore events, documents, the sequence of actions and the reasons behind the decision.
Avoid “Transparent pricing is important to you, isn't it?” It suggests the desired answer. A more neutral question is “How did you compare the costs of the offers, and what was unclear?”
Open questions and specific stories are recommended in the GOV.UK guide to in-depth interviews. Let participants explain their reasoning; do not turn the conversation into an agency presentation.
Begin by explaining the purpose of the conversation and how the answers will be used. If you need a recording, agree this with the participant in advance. You could allocate 30–45 minutes for the first version of your guide; this is a working estimate, not a required duration.
Follow up on broad answers: “Can you give an example?”, “What happened next?”, “At what stage did you learn that?” Do not ask participants to disclose another party's confidential documents or information unnecessary for the research.
For each observation, keep a brief record of:
Look for both similarities and contradictions. If one person complains about a long form while another considers a detailed brief helpful, investigate differences in their needs and stage of selection. Do not average the answers into a statement that describes nobody.
Keep access to interviews and collected material restricted. Agree separately on quotations intended for a public case study or website; an internal conversation does not grant permission to publish the customer's name.
Imagine that several participants in one segment tried to establish who would write the website copy and how many revision rounds were included before contacting a supplier. This is an illustrative situation, not a finding from research with the agency's customers.
The hypothesis is that the service page lacks an explanation of responsibilities and approval stages. A possible change is a short section covering materials, revisions and the next step after an enquiry.
First, check whether people from the target audience understand the section. After implementation, monitor recurring questions and enquiry quality. Interviews alone cannot justify promising a specific increase in sales.
A useful output includes a description of the segments studied, observations with supporting evidence, an outline of customer questions and prioritised hypotheses. Each hypothesis should relate to a specific decision: page content, an advertising argument or the process for initial contact.
These materials help shape a service page and a 90-day marketing plan.
When commissioning market research and marketing strategy, agree whether interviews are needed, who should be invited and how the findings will be delivered. Participant recruitment and answer analysis should be included in the scope before research begins.