KjerrCustomer feedback analysis,
step by step.
Customer feedback analysis turns a pile of requests into a short list of things worth doing: group what people said, count who said it, weigh it by who they are, and check what you did about it. This guide walks through five steps and the reports that do the counting in Kjerr.
What customer feedback analysis is, and is not
There are two kinds of feedback, and they need different analysis. Survey answers, reviews and support transcripts are free text in bulk, and making sense of them means coding or sentiment analysis. Feature requests on a board are different: each one already has a title, votes, a topic, a status and a known customer. Analysing those is mostly careful counting.
This guide is about the second kind. It will not teach text mining. It covers what to count, in what order, and how to avoid the two usual mistakes, which are counting duplicates twice and treating every voter as equal.
How to analyze customer feedback in five steps
- Put it in one place. Analysis of five inboxes is five analyses. Bring board posts, requests from support and imports into one list. The guide on how to collect customer feedback covers this step.
- Merge the duplicates. Ten requests for the same thing with three votes each look smaller than one request with thirty. Merge before you count. Kjerr suggests similar ideas while someone types, and AI merge finds the ones that slipped through for a teammate to approve.
- Group by topic. Give every idea a topic, then look at the share each topic holds. A topic that keeps growing is a theme, whatever the individual requests say. On a long board, let Insights do the pooling: it reads the ideas and sorts them into observations and themes for you to check.
- Weigh by who is asking. Thirty votes from trial accounts and thirty from your largest plan are not the same signal. Look at company, plan and MRR beside the votes.
- Check the response. Count how long requests take to ship and how many still have no answer. Analysis that ends at a ranking forgets the people who asked; the feedback loop is not closed until they hear back.
Customer feedback analytics worth watching
Customer feedback analytics do not need to be many. These three, compared with the period before, cover most decisions.
Volume and source
How much feedback arrived, per day, week or month, and whether it came from the board, from a teammate posting for a customer or from an import.
Time to ship
How long a request takes from posted to shipped. The median says what is typical; the slowest requests say what customers remember.
Awaiting response
How many ideas have no status and no reply, and for how many days. This is the number that tells you whether people are being answered.
A customer feedback dashboard, built in
A customer feedback dashboard should answer a question and then lead to the requests behind the answer. In Kjerr that page is Reports, and it needs no setup.
- New feedback over time, with the period's total and the previous period for comparison.
- Ranges from seven days to twelve months, or custom dates. Three months is the default.
- Breakdowns by status, by topic and by source.
- Top submitters, and top responders on the team.
- Click any bar and the inbox opens on exactly those ideas.
- Every chart exports as CSV and PNG.
Seeing topics at a glance
Grouping is the step that turns a list into themes, so the board makes it visible to the team and, if you choose, to visitors.
- Topics can be colour-coded from 52 colours, picked by hand or assigned automatically.
- The Topic map is an optional board section: a chart where each topic's slice grows with its share of ideas.
- The largest eight topics get their own slice; the rest fold into Other topics.
- Feedback by topic in Reports uses the same colours.
- Users & Companies filters people by plan, company and activity, and saves the segment.
Insights: the grouping, done for you
Steps two and three are where a long board costs the most time, so Kjerr has a switch for them. Turn on Insights and AI reads every idea and the comments on it, pools the ones that ask for the same thing into observations, keeps what people praise as strengths, and files both under themes that take your topics' names where they fit. Each is ranked by reach: the distinct people behind it, authors, voters and commenters counted once each. Every count opens the ideas and the sentences it came from, so the analysis is checkable, not a summary to take on trust. The details are on the feedback analysis tool page.
If you would rather ask your own assistant, Claude, ChatGPT, Cursor and other clients read a board through the MCP server, with read-only access to ideas, votes, comments and counts. Setup is in the MCP server documentation.
What Kjerr's reports leave out
The reports count what a feedback board holds. These are deliberately not included:
- Sentiment analysis or text mining.
- Automatic tagging on the board, or AI-written replies.
- NPS, CSAT or survey scores.
- A priority score. Scoring is the internal roadmap's job, with votes and MRR beside each RICE score.
- Custom dashboards or a chart builder.
Questions, answered plainly.
What is customer feedback analysis?
Customer feedback analysis is the process of turning raw feedback into decisions: grouping what people said, counting how many said it, weighing it by who they are, and checking what was done about it. The output is a short list of problems worth solving, with evidence.
What is the difference between feedback analysis and feedback analytics?
Analysis is the thinking; analytics are the numbers it rests on. Customer feedback analytics are the counts and trends, such as feedback per week, by topic or by source. Analysis is reading those numbers next to the requests themselves and deciding.
Which customer feedback metrics matter most?
Three: how much feedback arrives and from where, how long a request takes to ship, and how many requests are still waiting for any answer. The third is the one customers feel.
What is a customer feedback dashboard?
A customer feedback dashboard is one page that shows those numbers for a period and lets you open the requests behind each one. In Kjerr it is the Reports page.
Do I need sentiment analysis or AI to analyze feedback?
Not sentiment analysis: it earns its keep on thousands of survey answers or reviews, and a board of requests is already structured, with votes, a topic, a status and a customer on each. AI helps with one step, the grouping. Kjerr's Insights pools ideas that ask for the same thing and ranks the groups by the people behind them, with every idea and quote one click away.
How often should feedback be analyzed?
Read new feedback as it arrives and look at the numbers on a fixed rhythm, monthly for most teams. Compare each period with the one before it; a single month says little.
Related pages
Give your users an answer.
Kjerr is free during early access, with every feature included and no credit card. When paid plans arrive they will be priced per workspace, never per admin seat and never per tracked user, so the number of people who post, vote or comment never changes the bill.