At my previous software company, we ran a weekly "Voice of the Customer" meeting. The customer success team went through every reason customers gave for leaving that week, with someone from every department in the room.
It worked. But it took hours of reading.
Before I wrote Feed a Starving Crowd, I did the same thing by hand on a bigger scale: I read the three-star reviews of the top 100 marketing books on Amazon and wrote down every complaint. That list sat next to me the whole time I wrote the book.
This tool does the first pass of that work in seconds.
How do you use the Pain-Point Analyser?
- Describe what you sell in one line. It helps the AI understand the feedback.
- Paste your feedback. Reviews, support tickets, emails, survey answers, exit reasons, sales call notes. Put a blank line between entries. Up to 20,000 characters.
- Read the instant analysis. It runs in your browser, so nothing leaves your computer: theme counts, the share of entries mentioning each theme, quotes in customers' own words, and the phrases that keep coming up.
- Click Analyze with AI for the deeper read: pain points ranked with representative quotes, jobs-to-be-done, churn risks, pricing and willingness-to-pay signals, and five recommendations.
Remove names, emails and phone numbers first. When you use the AI, your text is sent to an AI model to analyze and isn't stored by this site.
How the instant analysis works
The browser pass splits your text into entries, then checks each entry against keyword groups for eight themes: price and value, cancelling and leaving, speed and delivery, bugs, support, ease and confusion, wrong fit (amount, timing, options) and praise.
Theme share = Entries that mention the theme ÷ Total entries × 100
It also counts the words and two-word phrases that repeat, leaving out common filler words, and gives a rough tone from positive and negative keywords. It's a keyword count, not a judgment, so always read the quotes.
A worked example: a coffee bean subscription
The sample is twelve made-up comments about a coffee subscription. The instant pass finds speed and delivery in 42% of entries, wrong fit (amount and frequency) in 33% and price in 25%, with mixed tone overall.
Look at what's underneath. "Too expensive for two of us drinking three cups a day" isn't really a price complaint. It's a fit complaint: the household needs a bigger bag at a lower price per gram. "Every two weeks is too often and every four is not enough" is the same thing. And two customers mention pausing or cancelling, which makes them churn risks you can call this week.
That's the shift the AI analysis makes for you: from "what did people say" to "what job were they trying to get done, and what do we change."
How do you read the results?
- One theme in more than a third of entries is a pattern, not noise. Fix it before anything else.
- Price complaints are often fit or value complaints. Check whether customers want a cheaper plan or a different one.
- Any mention of cancelling or pausing is a churn signal. Contact those customers, and make pausing easy.
- Praise tells you what not to break when you fix the rest, and gives you copy for your website in your customers' words.
Feedback tools can't tell you how many silent customers feel the same way. Use the themes to decide which questions to ask next, then measure.
What to do next
- Turn the top themes into fixes with how to reduce churn and the Churn Risk Checklist.
- Collect better feedback at the exit with these exit survey questions.
- If price keeps coming up, test it properly with the willingness-to-pay guide.
- Looking for a new market? Use the three-star review method in how to find a niche market.
Frequently asked questions
Is my customer feedback kept private?
The instant analysis runs entirely in your browser, so that text never leaves your computer. If you click Analyze with AI, the text is sent to an AI model to analyze and isn't stored by this site. Remove names, email addresses, phone numbers and anything else that identifies a person before you paste. Your text is saved only in your own browser so you can come back to it.
How much feedback do I need?
A dozen entries is enough to see the first themes, and fifty to a hundred gives you a much more reliable picture. Mix sources where you can: reviews, support tickets, exit survey answers and sales notes often reveal different problems. The tool accepts up to 20,000 characters at a time, so for bigger piles, analyze one month or one source at a time.
What is a job-to-be-done?
It's the progress a customer is trying to make when they buy from you, described in their terms rather than yours. A coffee subscriber's job might be 'great coffee at home every morning without thinking about it.' Framing feedback as jobs helps you see that a complaint about price or delivery is really about the job not getting done, which points to better fixes.
Where should I find feedback to paste?
Start with what you already have: support emails, online reviews, cancellation reasons and survey answers. For research on a new market, use three-star reviews of competitors, app store reviews, Reddit threads and video comments. Three-star reviews are especially useful because the reviewer usually liked parts of the product and explains clearly why it fell short.
Built on The Subscription Playbook
This tool turns frameworks from Robert Coorey's book into something you can use in minutes. Robert is the author of The Subscription Playbook and a Director at BPAG Accountants. More about Robert.