Customer Interview or Observation: What Each Can Reveal

The common mistake with a customer interview or observation is treating one as a shortcut to the other. Ask someone to describe a recent experience and you may hear a polished explanation; watch them attempt the task and you may see a workaround they never thought to mention. Choose the wrong method and you can spend time answering a question it was never equipped to settle.

These methods answer different questions. Interviews reveal how people interpret and remember an experience. Observation reveals what they do in a particular setting. Neither gives direct access to every motive, and neither is a verdict on what all customers want.

Match the method

Start by writing the question you need to answer in plain language. “Why did the customer abandon the booking?” is not the same as “Where does the customer hesitate while booking?” The first calls for an account of someone’s reasoning; the second calls for attention to behaviour at the point of use.

Use an interview to understand a person’s vocabulary, priorities, perceptions or recollection. Use observation to understand sequence, friction, workarounds or interaction with a physical or digital environment. If your question is about what people think, remember or value, begin by asking. If it is about what happens when they try, begin by watching.

Keep the question narrow enough to guide the session. A small business considering a new appointment process might want to know whether customers understand the cancellation wording. That is more useful than asking whether they like the whole booking experience, which can invite broad opinions without identifying a decision.

Ask for meaning

An interview gives a customer room to describe what mattered, what they expected and how they made sense of an event. It can uncover terms your business does not use, concerns that are invisible in a screen recording, and differences in how people define a successful result.

Ask about a specific, recent occasion rather than inviting predictions. “Tell me about the last time you tried to reschedule” is more grounded than “Would you use a rescheduling feature?” Follow up with “What happened next?” or “What were you trying to do at that point?” Then allow a pause; suggesting an answer too quickly can steer the account.

Interviews are particularly helpful when an experience is private, spread over time or difficult to witness. A customer can explain how they compared providers before making contact, or what they hoped to avoid when choosing a service. Their explanation is still a recollection, not a recording of the decision as it happened.

Keep your assumptions out of the question. “Was the form confusing?” points towards a particular answer. “How did you decide what to enter here?” leaves room for the customer to describe their approach. Record the person’s words accurately and separate direct statements from your interpretation.

Watch behaviour

Observation helps when the task itself matters: how someone finds a product, follows instructions, uses a service counter or moves through a booking page. It can expose pauses, repeated attempts, detours and reliance on another person. These details may be hard to recall later, even for a thoughtful customer.

Give the participant a realistic task and make the setting as ordinary as practical. For a digital task, ask them to find an available appointment and stop before confirming it. For a physical service, watch how they locate the right queue or identify what information they need at reception. Avoid coaching unless help is part of the experience being examined.

Note what happened before assigning a cause. “Returned to the previous page, then opened the price details” is an observation. “Did not trust the price” is an interpretation that needs checking. A quiet participant may be concentrating, unsure, distracted or simply not inclined to speak aloud; visible behaviour rarely explains itself.

Observation has limits. Someone who knows they are being watched may behave differently, and one session cannot tell you how common a behaviour is. You see actions in a particular task and context, not a complete portrait of the customer or their private reasoning.

Understand the limits

An interview can miss behaviour. People forget small steps, compress a complicated event into a neat story or describe what they think they ought to have done. Their account still matters: their interpretation may itself be important. But an interview alone cannot establish the exact sequence of actions.

Observation can miss meaning and context. You may see someone abandon a form without knowing whether they lacked a document, changed their mind or were interrupted. Watching a task also cannot reliably reveal what they would do in another setting or how they feel about a longer relationship with the business.

Session design can distort either method. A leading prompt affects an interview; an artificial task affects observation. Keep the activity close to a real customer situation, and make claims no broader than the evidence permits.

Plan the session

For an interview, prepare a short set of open prompts rather than a rigid script. Begin with the customer’s recent experience, then ask about decisions, obstacles and what happened next. Leave room for an unexpected detail, and avoid turning the conversation into a pitch for your business.

For observation, define the task, setting and notes you will capture. Record the order of actions, visible hesitation, tools used and places where the participant changes course. If appropriate and they agree, invite them to describe what they are thinking, while remembering that speaking during a task can alter how they approach it.

Say you run a neighbourhood repair shop and want to understand why customers arrive without complete booking details. You ask one customer about a recent booking; they say they thought the equipment model was optional. You then watch another customer book and see them reach the model field but leave the page to look for the appliance label. The first account points towards clarifying the wording; the second suggests helping customers locate the information. The trade-off is clear: interviews reveal the customer’s interpretation, while observation shows where the task becomes difficult. These are invented possibilities, not evidence about a particular shop.

Combine with care

Use both methods when you need to compare what someone says with what happens during a relevant task. A short interview can establish their aim and recent context; observation can show their approach; a follow-up question can help explain a moment you noticed. This creates a fuller account without making the two kinds of evidence interchangeable.

Keep notes in separate columns or sections: “said”, “did” and “our interpretation”. That separation makes it harder to turn a comment into a behavioural fact or a visible action into a claimed motive. When an account and an observed action differ, keep the difference visible and ask a neutral follow-up rather than deciding one must be wrong.

You do not need both methods for every small decision. If the question concerns customer language, interviews may be enough to improve a message. If it concerns a visible bottleneck, observation may be the more direct starting point. A second method is useful when it answers a remaining question, not simply because collecting more material feels thorough.

Turn evidence into action

Before sessions begin, name the business decision the findings might inform. It could be whether to revise a booking prompt, change how staff explain a service or investigate a particular obstacle further. This keeps the work connected to an action while leaving room for the answer to be “we need to understand this better”.

Afterwards, sort notes by the question, not by which customer made the most memorable comment. Look for recurring kinds of difficulty, contrasting interpretations and observations that challenge your initial assumption. Repetition is a reason to consider a pattern, not proof that it applies to every customer.

For each proposed change, record the evidence behind it and what remains uncertain. If a decision would affect a broad customer group, consider whether the sessions covered the people and circumstances relevant to it. A focused conversation can improve a local detail; it cannot stand in for every form of customer feedback or establish how widespread an issue is.

Which method fits your question?

Skip unnecessary polish

Do not spend time building an elaborate interview script, buying recording equipment or creating a formal research report before you know what decision needs support. A clear question, a suitable setting, careful notes and respectful consent matter more than presentation. The point is not to collect impressive transcripts; it is to understand enough about a customer experience to make a considered next choice.

For a customer interview or observation, ask when meaning is missing, watch when behaviour is unclear, and combine the methods when you need to connect the two. Keep your claims close to what people said or did, and let the next decision reflect both the evidence and its limits.