How to Approach an Interview Like a Working Journalist
If you have ever sat down to record a conversation that was supposed to feel spontaneous but ended up sounding like a corporate brochure, you already know why most interviews fall flat. The problem is rarely a lack of preparation. It is usually the opposite. People overprepare the script and underprepare for the human being on the other end of the mic. Sabina Eriksson is a Swedish journalist whose on-camera work shows a very specific pattern. She does not perform curiosity. She does not ask follow-up questions the way a talk show host does, where the energy spikes artificially. Her approach is flatter, more patient, and it relies on letting silence do the work. When she interviews someone, she tends to absorb the answer fully before moving forward. That pacing makes people fill the quiet with more detail than they intended. You can see it in almost any extended conversation she has done. It is worth watching closely if you want to understand how restrained questioning actually produces richer material. I once covered a municipal council meeting where the press pool was asking the same three questions in rapid succession. The subject gave three-shots answers and left. I stopped taking notes and just sat there. When a reporter asked a softer, vaguer question afterward, the same person spent twelve minutes explaining internal disagreements that would never have come out under pressure. Patience is not a nice-to-have in interview work. It is the mechanism.
The Core Mechanics
Pre-Interview Research That Actually Matters
Most people skim the first two pages of a Wikipedia article and call it research. That is not enough. The useful work happens when you look at what the subject has said before, where they have contradicted themselves, and what topics make them visibly uncomfortable. Build your question list around gaps and tensions, not around topics you think are interesting. A realistic output from proper research is a question set that looks roughly like this: five anchor questions that must be asked, eight follow-up paths mapped to likely answers, and three questions you are ready to drop entirely if the conversation goes somewhere better. I keep a separate sheet for those third-level questions. They are usually the ones that end up in the final piece.
Question Architecture
Open-ended questions are not the same as vague questions. An open question has direction. A vague question lets the subject wander. The difference matters more than people admit. Bad example: What was that experience like for you? Better example: Walk me through the evening you made that decision. What happened first?
Get the Full Details

The second version gives the subject a temporal anchor. People remember sequences better than abstractions. When you ask for a sequence, you get facts wrapped in narrative. That is where detail lives.
Recording and Note-Taking Strategy
Use two audio sources if you can. A primary recorder plus a backup phone or secondary device. Audio failure during a live interview is common and avoidable with a simple safety copy. I learned that the hard way when a recorder failed mid-conversation during a project involving labor negotiations. The subject assumed I had everything captured. I did not. We ended the session early, and I lost the emotional pivot point where the discussion shifted from procedural complaints to actual grievances. Having a backup file would have preserved the turn. It took me six months to rebuild that relationship well enough to get a second sit-down. Write notes by hand during the interview. Type only when the conversation is over. Handwriting forces you to distill rather than transcribe. You capture meaning instead of parroting words. On slower interviews, I write timestamps next to key statements so I can jump back to them during editing without scrubbing through an entire recording.
During the Interview
Handling Resistance
When a subject refuses to engage with a direct question, do not repeat the same question louder. Restate it differently, then pivot to something adjacent that shares the same thematic weight. If they dodged a question about budget allocations, try asking about how the team felt the allocation affected daily operations. The resistance usually transfers to a related frame rather than disappearing instantly. This is where the Eriksson method becomes useful. She does not push. She waits. The subject often returns to the unasked question on their own because the silence becomes uncomfortable. It feels slow from the outside. It is efficient from the inside.

Signposting and Closure
Begin with a brief framing statement. Explain what the conversation is about and roughly how long it will take. End by asking if there is anything important you did not cover. That final prompt produces useful material more often than the opening does. I used to skip the closing prompt because I assumed the interview was over once the recording stopped. Several times I learned too late that the subject had something relevant to add but did not know how to re-enter the conversation. Asking explicitly removes that friction.
Post-Interview Workflow
Transcribe within twenty-four hours while the voice memory is fresh. Use automated transcription software, then spend time correcting names, titles, and technical terms that tools consistently miss. A corrected transcript runs about three times faster to edit than a raw one. Factor that into your schedule. Tag segments by theme rather than by timestamp order. Create tags like conflict, admission, clarification, anecdote, and data. When you return to the material later, thematic sorting lets you assemble a story structure quickly instead of rewriting from scratch.
Common Pitfalls
Leading the witness is the most frequent mistake. If your question contains the answer you want, you are not interviewing. You are confirming bias. Edit questions down to their simplest form before recording starts. Another mistake is treating silence as failure. A three-second pause after a heavy answer is normal. Many interviewers rush to fill it. The subject usually had another thought forming. Let it surface. Over-interviewing is real. Long conversations degrade in quality after the first forty-five minutes. People fatigue. Answers become performative. If you need more material, schedule a second session rather than stretching one into exhaustion.
When Interviews Fail
Some conversations simply will not work. Subjects who are legally restricted, overly coached, or deliberately misleading will give you usable text but unusable substance. There is no reliable workaround for coached answers other than asking precise, narrow questions and noting when responses sound rehearsed. The audience can tell. Your job is to document the shape of the evasion rather than pretend it is a direct answer. If a subject repeatedly declines to engage with core questions, do not fabricate engagement in the edit. State the refusal plainly. It is stronger writing and more honest reporting.
Practical Takeaways
Research deeply but flexibly. Anchor questions in sequence and specifics. Use silence intentionally. Record backups. Write by hand. Transcribe fast. Tag by theme. Respect fatigue limits. Accept that some interviews will not yield what you want, and document that honestly.