Feedback Analyzer
Weekly themes from every survey, ticket and review.
Feedback Analyzer
Customer Support & Success
- RuleDone:
Each week it collects new items since the last run.
- RuleDone:
It cleans them, removes duplicates and strips personal data.
- AIDone:
AI labels themes, sentiment and type, and picks a quote.
- RuleDone:
Urgent items (injury, lawsuit, data breach…) alert someone at once.
- AIWaiting for approval:Waiting for you
Rules count, compare and pick quotes; AI writes short summaries.
- RuleNot started:
The report is published and theme tasks go to owners.
For example: A three-location dental clinic connects its Google reviews and a post-visit Typeform.
Waiting for you. Nothing has changed yet.
Standard. Low-risk, reversible changes. Every action is logged.
What is Feedback Analyzer?
In shortFeedback Analyzer is an AI automation for Customer Support & Success teams that works with Asana, Gmail, Google Business Profile, Google Docs, Google Forms and 4 more apps. Turns survey answers, support tickets and Google reviews into a weekly theme report with real counts, trends and quotes, sent to each theme's owner. It is built for product manager, head of customer experience, owner of a clinic, restaurant chain or service business and theme owners.
Last reviewed
Key facts
- Department
- Customer Support & Success
- Starts when
- Every week (default Monday 08:00), with an optional daily collection for high volume
- Works with
- Asana, Gmail, Google Business Profile, Google Docs, Google Forms, Jira, Slack, Typeform, Zendesk
- Run modes offered
- Shadow, Draft, Assisted and Auto
- Human control
- Standard
- Status
- Available now
The problem today
Someone reads hundreds of survey comments and reviews once a quarter, copies quotes into a deck and guesses the counts. Today: 1–2 days per quarter, and most feedback is never read.
What it does
Every week it collects new feedback from Typeform, Google Forms, Zendesk and Google reviews, removes personal data, and labels each item with themes from your list, plus sentiment and type. Rules count the themes, compare with last week, and call something a trend only when the numbers support it. A short summary with real quotes goes to Slack, email or a Google Doc, and themes over your threshold become tasks for their owners. Safety or legal items alert a person right away.
How does Feedback Analyzer work?
How it works- Rule you set
- AI step
- A person decides
- Step 1Rule
Each week it collects new items since the last run.
- Step 2Rule
It cleans them, removes duplicates and strips personal data.
- Step 3AI
AI labels themes, sentiment and type, and picks a quote.
- Step 4Rule
Urgent items (injury, lawsuit, data breach…) alert someone at once.
- Step 5AI
Rules count, compare and pick quotes; AI writes short summaries.
- Step 6Rule
The report is published and theme tasks go to owners.
What you get
Key features- Starter list of 12 themes each for dental, clinic, restaurant, SaaS and other businesses; fully editable.
- New theme suggestions are grouped automatically, and the owner accepts or rejects them.
- Small-sample rule: "few items: not a trend" below 10 items or 50 in total; trends only when statistically significant.
- Real quotes only (exact text), 3 per top theme, from different customers and sources.
- Theme tasks in Slack, Jira or Asana at 10+ items or +50% vs last period.
- Health-term redaction mode for clinics; staff names are always redacted.
- Reads survey answers handed over by Survey Automation (28).
How much does it do on its own?
Run modesPick a mode to see what Feedback Analyzer would do at each step.
Every automation starts in Shadow. Move it up when you are ready, and back down at any time. How run modes work
Feedback Analyzer in Shadow
Runs on real data and shows what it would have done. No app is changed and no message is sent.
- Rule
Each week it collects new items since the last run.
Watches for it
- Rule
It cleans them, removes duplicates and strips personal data.
Runs on real data, read only
- AI
AI labels themes, sentiment and type, and picks a quote.
Runs on real data, read only
- Rule
Urgent items (injury, lawsuit, data breach…) alert someone at once.
Runs on real data, read only
- AI
Rules count, compare and pick quotes; AI writes short summaries.
Shown to you as a suggestion
- Rule
The report is published and theme tasks go to owners.
Shown to you as a suggestion
Shadow vs Auto for Feedback Analyzer
Auto is the most it can do on its own. Here is each step side by side.
| Step | Shadow | Auto |
|---|---|---|
| In short | Runs on real data and shows what it would have done. No app is changed and no message is sent. | Runs without approval, with limits, alerts and a kill switch. Unlocks after a trial period in Assisted. |
| Step 1Each week it collects new items since the last run. | Watches for it | Runs on its own within your limits |
| Step 2It cleans them, removes duplicates and strips personal data. | Runs on real data, read only | Runs on its own within your limits |
| Step 3AI labels themes, sentiment and type, and picks a quote. | Runs on real data, read only | Runs on its own within your limits |
| Step 4Urgent items (injury, lawsuit, data breach…) alert someone at once. | Runs on real data, read only | Runs on its own within your limits |
| Step 5Rules count, compare and pick quotes; AI writes short summaries. | Shown to you as a suggestion | Runs on its own within your limits |
| Step 6The report is published and theme tasks go to owners. | Shown to you as a suggestion | Runs on its own within your limits |
| True in every mode | Never replies to customers; every action is internal. | |
What it will never do on its own
Safety & human control- Never replies to customers; every action is internal.
- Removes emails, phones, order and card numbers, links, customer names and staff names before AI and before storage.
- Never ranks or names staff; no "customers think" from fewer than 10 items, and "most customers" only at 50% or more.
- Clinics must use a BAA-covered deployment or turn on health-term redaction.
- Changes to the theme list always need the owner.
- You approve before it acts
- Your data stays yours
- It knows its limits
- Stop anytime
What will you see on your dashboard?
Results you can trackTargets are quality goals the automation is built to reach. Estimates are calculated from your own volumes.
- Tracked
Items analysed vs manual baseline
Measured on your own data
- Tracked
Top themes trend
Measured on your own data
- Tracked
Tasks created and closed per theme
Measured on your own data
- Tracked
Unclassified share
Measured on your own data
- Target
Time from feedback to "seen by owner" (target: under 7 days)
A goal it is built to reach
- Target
Theme agreement with a human (target: ≥ 85%)
A goal it is built to reach
- Target
Sentiment accuracy (target: ≥ 85%)
A goal it is built to reach
- Target
Urgent recall (target: 100%)
A goal it is built to reach
- Target
Exact quotes (target: 100%)
A goal it is built to reach
- Target
Personal data in reports (target: 0)
A goal it is built to reach
A three-location dental clinic connects its Google reviews and a post-visit Typeform. Monday's report shows "wait time" rising, with three real quotes, and a Slack task goes to the front-desk manager. A review mentioning an injury alerts the practice owner the same day.
Who is Feedback Analyzer for?
- Product manager
- Head of customer experience
- Owner of a clinic
- Restaurant chain or service business
- Theme owners
Industries: Healthcare & clinics, Hospitality & restaurants, SaaS & tech, Professional services
Also useful for: Analytics & Reporting
Connects to the apps you already use
Works withApp names are trademarks of their owners. Charviam connects to them; it is not a partner or reseller.
- Asana(optional)
- Gmail
- Google Business Profile
- Google Docs(optional)
- Google Forms
- Jira(optional)
- Slack
- Typeform
- Zendesk
Better together
Automations that hand work to Feedback Analyzer, or pick up where it stops.
What do you need to set it up?
Setup- Connect sources (read only): Typeform forms, Google Forms, Zendesk, Google Business Profile locations.
- Choose outputs (Slack, internal email, Google Doc template) and a task tool (Slack, Jira or Asana).
- Edit the theme list and owners; set urgent keywords and an urgent contact (required).
- Add staff names for redaction and internal report recipients.
- Clinics: confirm a BAA-covered deployment or turn on health redaction.
Typical setup
- 01Connect your apps
- 02Configure the rules
- 03Run a watch-only trial
- 04Go live
Not included
- Replying to customers or reviews (that is 13) and sending surveys (that is 28).
- Judging individual staff performance.
- Statistical claims about the whole customer base.
Questions about Feedback Analyzer
FAQSee Feedback Analyzer on your own data
WalkthroughIt starts in Shadow, where it only watches and suggests, so nothing changes in your apps until you say so.