Customer interviews are for gathering data for analysis and interpretation. If you’re summarizing or paraphrasing notes, you’re wasting your time and risking making some critical mistakes.
Over my career, I’ve been involved in at least a dozen large scale research projects with other researchers. At the end of the process, we’d gather our notes together to synthesize them. We were shocked at how different the notes were.
My notes were lengthy, and included stop words and pauses. My coworker Keith’s were paraphrased and included conclusions he’d come to while listening to the customer. I was recording, Keith was summarizing. While I was impressed with his ability to think AND listen at the same time, his attempts to synthesize at the same time likely robbed him of insights.
While my notes took time to go back and synthesize after the fact, it was also an accurate representation of what the customer actually said, rather than my interpretation of such.
Stakeholders like pull-quotes: What did the customer actually say. Summary notes carry less credibility and could be misleading.
Summary notes also mean you’re deciding in real time what’s important and what’s not. As you’re doing research, one unusual response is an anomaly: ten may be a sign of a pattern, an unanticipated problem that needs attention. If you’ve neglected to take note of it in your original transcript, you can’t go back.
When Martin.ai reviews interview transcripts, he’s looking for subtle cues to interpret customer intent. The difference between “often” “frequently” and “all the time”. Paraphrasing notes introduces human bias and error. In the past, recording, transcribing and analyzing was time-consuming and tedious. Martin.ai was designed to assist in the analysis phase so even busy researchers can get quality analysis.
- Consider your interviewing to have two phases: recording and synthesizing. Don’t do them at once.
- Have a note-taker with you so you can facilitate the conversation and they can be capturing the notes. Recording your interviews is ideal to be sure you’re capturing everything.
Originally published on Martin.ai in December, 2015