Sentiment breakdown
Every review is scored positive, neutral or negative — not just "good or bad," a genuine three-way read, shown as an overall percentage and a per-category split.
Category frequency
Reviews are grouped into the categories reviewers actually talk about — pricing, support, quality, shipping and more — sized by how many reviews mention each one.
Severity, not just sentiment
Negative reviews are further scored from minor annoyance to dealbreaker, so a frequent small gripe and a rare deal-ending complaint don't look the same on the page.
Explicit requests & implied needs
Reviews that directly ask for a feature are separated from reviews that imply a gap without asking for anything by name — both count as opportunity signal.
Opportunity matrix
Opportunities are plotted by frequency and severity, so you can see at a glance which weakness is both common and serious enough to be worth attacking first.
Real quotes, not summaries
Every claim in the report is backed by an actual review quote you can click through to, sorted by how extreme and how confidently-classified it is.
One-click PDF export
The full report — sentiment, pain map, opportunities, quotes — exports as a real, selectable PDF you can share or file away, generated directly in the browser.
Chrome extension
Runs directly on the page you're already looking at — no copying review text into another tool, no spreadsheets.
Built on evidence, not estimation
The thing every feature above has in common: nothing is guessed. Categories, sentiment and severity come from a typed decision model judging one review at a time; every count and percentage is arithmetic over those judgments, computed in code. The only thing a language model is asked to write is the plain-English explanation of numbers that were already decided.
See it on a real page
Install the extension and run it on any competitor's review page — no account required.