Choose one learning goal
Write down what the team wants to understand before drafting the questions. For example: “We want to learn where new account owners get stuck when inviting teammates.” This gives you a way to decide which follow-ups matter and which belong in another conversation.
Invite someone who has recently done the relevant task. Tell them what you will cover, how much time you need, and whether you will use transcription or an AI assistant. Ask them to choose examples they are comfortable sharing, with sensitive customer or company information removed where appropriate.
A 30-minute agenda
Keep the introduction short and leave most of the time for the customer's account. If you also need to show a proposed feature, do it after discussing their current experience so the design does not set the direction of every answer.
- 0–5 minutes: explain the learning goal, confirm the plan for the call, and ask about their role.
- 5–18 minutes: walk through the last time they performed the task, including tools and workarounds.
- 18–25 minutes: explore the most consequential difficulty and what they tried to do about it.
- 25–30 minutes: reflect back what you heard, invite corrections, and agree on any follow-up.
Questions that give the conversation somewhere to go
Start with an event, then follow its sequence. When the customer describes a broad frustration, ask for a recent example. When they request a feature, ask what they would use it to accomplish. The request is useful; so is the job behind it.
- When did you last do this? What were you trying to get done?
- What did you do first? What happened next?
- You mentioned that step was frustrating. Can you walk me through an example?
- How did you finish the task? Who else became involved?
- If that part went smoothly, what would change in your work?
Keep observation and interpretation separate
After the call, write down the observed behavior or the customer's words, followed by your interpretation and the question to investigate next. For example: “The customer exported a list and emailed it to a colleague” is an observation. “They need a collaboration dashboard” is a product hypothesis. Keep both, with different labels.
Do not turn one conversation into a claim about every customer. Compare it with other relevant conversations and product evidence. If the team wants to build something, specify which assumption a prototype or small change would test.
Using Q during the conversation
In SideQuest, you can ask Q for a follow-up or a recap of the shared discussion. Check that recap with the customer, especially the parts that imply a priority or commitment. Q can help organize the conversation; the customer remains the source for their experience. Close by naming what you will follow up on, without promising a feature or date the team has not agreed to deliver.