Most feedback tools count votes on a public board. Sentriment reads everything instead, including the feedback nobody ever posts: support conversations, app store reviews, in-app messages, survey responses and imported spreadsheets.
Every item is analyzed for sentiment, emotion, type and the specific aspects it mentions, then grouped into named themes by meaning rather than keywords. "The app crashes when I pay" and "payment screen freezes" land in the same theme without anyone building a taxonomy.
Each theme carries three numbers: how many distinct people raised it, whether it is growing or fading, and the total monthly revenue of the accounts behind it. You can also ask questions in plain language, such as what customers think of your pricing, and get a verdict with counts and the quotes that support it.
Identified users receive a 0 to 100 health score built from their own feedback history, so accounts drifting toward cancellation surface before they leave. Scores can be pushed back into Mixpanel or Amplitude as user properties.
Data is hosted in the EU (Frankfurt). Personal details are removed from feedback before anything is stored, and every AI output records the model and prompt version that produced it, so any claim can be traced back to the items behind it.
Pricing is per feedback item analyzed rather than per seat, so the whole team can read the analysis without adding cost. The free plan covers 100 items a month, and the first 1,000 imported items are free on every plan.