Turning feedback into themes
Two thousand pieces of feedback don't tell you anything until they become twelve themes. The problem is that theming by hand is slow, subjective and stops the moment the person doing it gets pulled onto something else.
In Lesson 03 of Voice of the Customer 101, we walk through how to go from raw feedback to themes you can act on: what a good theme looks like, how to avoid the trap of themes so broad they're useless, how to separate what customers say from what they mean, and where AI helps and where it still gets it wrong.
What is said in this lesson
The spoken text from the video's captions, split into paragraphs. Read it in two minutes or search it for the part you need; the captions are automatic, so a word may be off here and there.
Show the transcript
Hundreds of comments, one spreadsheet, and a meeting on Friday where someone asks, "So, what are customers actually saying?" The answer is not a word cloud. It is a short list of themes, each with proof behind it. Today, how to get from raw feedback to themes you can act on. Here is the framework. Four steps. Tag, group, name, nest. Tag each piece of feedback with what it is about in the customer's language. Can't export. not the name of the team that owns exports. Group tags that mean the same thing.
Name each group as a problem customers have. Then nest broad themes on top, specific sub themes underneath. The step most teams get wrong is grouping. Listen to these three. I can't find where to download my report. The spreadsheet export keeps timing out. Why does the file lose my filters? Different words. Same problem. Customers are struggling to get their data out. Now the reverse. Your reporting is fantastic. Same word report, but it is praise, not a problem. Group by keyword, and you get both wrong. Now nest them.
A theme is the customer problem at a level one leader can own. Getting data out, getting started, understanding my bill. A subthe is the specific version a team can fix. under getting data out. Can't find export. Export fails. Export loses settings. Why two levels? A leader decides with the theme. The team doing the work needs the sub theme. Keep two more things separate from the theme. The type, whether it is a complaint, a request, praise, or a question, and the sentiment. That structure is your taxonomy.
Keep it simple. Two levels. No more a theme list short enough to fit on one screen. Every sub theme belongs to exactly one theme. Give each theme a oneline definition. So two people tagging the same comment land in the same place. A small other bucket. And when it grows, that is a new theme knocking. And review the list once a month because your product changes and so do the problems. Three mistakes to avoid. First, grouping by keyword. It splits one problem into five and merges praise with complaints.
Second, naming themes after a solution. Add bulk export locks you into one fix. Getting data out lets the team find the best one. Third, letting the list sprawl. Every new tag becomes a theme and 6 months later, nobody trusts the chart. Merge before you add. Here is your homework. Take the feedback you gathered after lesson one. Tag each piece with a short phrase in the customer's meaning. Group the tags into themes that fit on one screen. Name each theme as a problem. Give each one two or three sub themes and one real quote.
That page is your first taxonomy. Or customer intelligence explained is presented by Hyperorbit, a gentic customer intelligence. See your app reviews grouped into themes with the free app review analyzer at hyperorbit.ai.
Every signal, on one orbit
Connect your first source in an afternoon. The first pass lands before your next standup.
Book a demo