Understanding Actual Workflows through Discovery Interviews
A discovery interview, as the source presents it, is a conversation of about thirty minutes aimed at learning how work is actually done.
Why Ask How Work Really Gets Done
Most people who are asked how their work gets done will describe the version they believe they should follow. The steps they actually take, the shortcuts they rely on, and the workarounds they have quietly built are often left out. A discovery interview is a method for closing that gap. As the source presents it, a discovery interview is a conversation of about thirty minutes aimed at learning how work is actually done.
The word discovery matters. It signals that the interviewer is not arriving with a finished answer. The conversation exists so that the person running it can later reason about the real workflow instead of guessing at it. Keep that purpose in mind as we look at the shape of a single session, because the rest of the method makes more sense once you can see how one conversation unfolds.
A Bounded Thirty-Minute Conversation
A discovery interview is a conversation of about thirty minutes aimed at learning how work is actually done. The length is fixed and short, and the conversation is aimed at a specific product or piece of work.
The short, fixed length is a design feature, not an incidental detail. Because the session is only about thirty minutes, the interviewer cannot attempt full coverage of everything the interviewee does. Instead, the interviewer focuses on one concrete instance of the work. A single real case, such as the last time a particular task was carried out, gives the conversation something specific to examine.
What do you think happens?
If a session can only last about thirty minutes, what should the interviewer aim to cover?
Reveal answer
One concrete instance of the work, rather than attempting full coverage.
The short fixed length pushes the interviewer to focus on one concrete instance of the work instead of trying to cover everything.
Tracing One Session from Opening to Notes
The source describes a sketch of one thirty-minute session, with the interviewer's internal annotations carried alongside the conversation. The details of that sketch are not reproduced here, so the trace below is an illustrative one written to show the reasoning pattern. Treat it as one way a session could move, not as a record of a real interview.
Following a Session Step by Step
An interviewer is running a thirty-minute discovery session about how a team processes incoming requests. Trace how the session moves from the first question to recorded notes.
- 1Open with a question about a recent task
The interviewer asks the interviewee to describe a recent request they handled. The question is anchored to a real instance rather than a general description of the job.
- 2Let the interviewee walk through one case
The interviewee explains what happened in that one case. Because the session is short, the interviewer keeps the focus on this single instance.
- 3Ask what happened next
The interviewer follows up on each step the interviewee mentions, so that the account of the work stays concrete and sequential.
- 4Annotate each answer as it arrives
The interviewer writes private notes beside the exchange. An answer that points to a real step is marked as a step. An answer that sounds like a workaround is flagged. An answer that raises a question is noted for follow-up.
- 5Record the open questions
Questions that were not resolved in the session are written down. Recording them now lets later follow-up build directly on what was learned.
- 6Hold the notes as one record
The result is the spoken conversation together with the interviewer's annotations and open questions, which together form the recorded notes.
The session ends with recorded notes that hold both what was said and what the interviewer made of it, with open questions saved for the next conversation.
Keeping Conversation and Annotations Apart
During a discovery session there are two layers of record. The first is the conversation, which captures what was said. The second is the set of internal annotations, which are the interviewer's own working notes. An annotation sits next to the conversation and is written for the interviewer, not for the person being interviewed. In the source's words, the conversation records what was said, while the annotations record what the interviewer made of it.
The two layers relate in a specific way. Each annotation refers back to a moment in the conversation, but the annotation adds a judgement about it. A raw transcript of questions and answers tells you what happened. Annotations let you mark which answers point to a real step in the workflow, which sound like a workaround, and which raise a question to follow up on. Without that second layer, a thirty-minute conversation becomes a pile of statements that is hard to reason about afterward.
Mistakes When Running a Discovery Interview
Common mistake Arriving with a finished answer
What it looks likeThe interviewer already believes how the work is done and asks questions only to confirm it.
Why it failsThe word discovery signals that the conversation exists to learn how people actually do their work. A predetermined answer leaves no room for the real workflow to appear.
Do this insteadTreat the session as a way to find out how the work is done, and let the interviewee's concrete account shape the next question.
Common mistake Trying to cover everything in one session
What it looks likeThe interviewer tries to ask about every task the team performs during the thirty minutes.
Why it failsThe short fixed length does not support full coverage. Spreading the time across many topics leaves each one shallow.
Do this insteadFocus on one concrete instance of the work and examine it closely.
Common mistake Writing only the transcript
What it looks likeThe notes contain a list of questions and answers with no marks on which answers matter.
Why it failsA raw record tells you what happened but not what it means for the workflow. Afterward, the notes are hard to reason about.
Do this insteadKeep annotations beside the conversation that mark real steps, workarounds, and open questions.
Common mistake Discarding open questions
What it looks likeAn unresolved point is not written down, and the next conversation starts from scratch on that topic.
Why it failsRecording open questions during the session is what lets later follow-up build directly on what was learned.
Practice: Sorting a Short Exchange
Read this short exchange from an illustrative session. Interviewer: Walk me through the last request you handled. Interviewee: I open the form, copy the details into the tracker by hand, and then I usually message the other team because the tracker never syncs. Sort the interviewee's statements into three kinds of annotation: a mark for a real workflow step, a flag for a workaround, and an open question to follow up. Then write one question you would ask next.
Hints
- Start with the interviewee's first statement. Is it a step in the work, or a question you need to revisit?
- The phrase about the tracker never syncing describes why a step exists. Consider whether that makes it a workaround.
- Your next question should follow up on one concrete instance, since the session is focused on one case.
Putting the Session Together
- 1A discovery interview is a conversation of about thirty minutes aimed at learning how work is actually done.
- 2The fixed, short length pushes the interviewer to focus on one concrete instance of the work rather than full coverage.
- 3Internal annotations are the interviewer's own working notes, kept separate from what the interviewee said.
- 4A session moves from opening questions through concrete follow-up to annotated answers and recorded notes, with open questions saved for later.
- 5Keeping the transcript and the annotations apart lets you later reason about the real workflow instead of guessing at it.
Key Takeaways
- A discovery interview is a conversation of about thirty minutes aimed at learning how work is actually done.
- The short, fixed length pushes the interviewer to focus on one concrete instance of the work.
- Internal annotations are the interviewer's own working notes, kept separate from the spoken conversation.
- A session moves from opening questions through concrete follow-up to annotated answers and recorded notes, with open questions saved for later.
- Keeping the transcript and annotations apart lets you reason about the real workflow afterward instead of guessing at it.