Every completed order asks for feedback. Every answer reaches the team that can act on it.
Post-purchase feedback usually dies in one of two places: nobody asks, or somebody asks and the replies pile up unread. This recipe closes both gaps. The moment an order is marked Completed, the customer gets a survey link. When they answer, AI reads the response, labels it, and pushes it straight to the team that owns it (Product, QA, or Support) while praise gets quietly set aside for your testimonials. Leadership gets the trend summary weekly without asking for it.
Free · no credit card · you connect your own apps
Installing provisions each one, wired together and ready to run.
Your order book, populated by Shopify or any platform or process you connect. Setting a row's status to Completed is what triggers the feedback invite.
The customer-facing form: star rating, comments, optional suggestion, optional email. Publishing it auto-creates the response sheet that everything else reads and writes to.
Runs daily. Finds newly completed orders, emails each customer the survey link, and marks the row so nobody gets asked twice.
Fires on every submission. Classifies sentiment, category, and theme, writes it back to the row, routes it to the owning team in Slack, and flags praise as testimonial-ready.
Runs Mondays at 8:00 AM UTC. AI summarizes recent responses into sentiment, category breakdown, recurring issues, and requested improvements, then emails leadership.
You connect your own accounts. Swap any app that offers an alternative at install.
Sends the automatic feedback-survey invite after an order is marked Completed.
Routes AI-classified feedback to the relevant team channel: Product for feature requests, Product/QA for quality issues, Support for complaints and delivery problems. Swap for Telegram or Discord.
Sends the weekly AI trend report to leadership.
Your store fills the Orders table: Shopify, another platform, or any process you connect. Once a day, the recipe sweeps that table for orders marked Completed that haven't been asked yet, emails each customer a link to the Order Feedback Survey, and flags the row so the same shopper is never invited twice.
The survey is short on purpose: a star rating, free-text comments, an optional suggestion, and an optional contact email. No login, no account, no friction. Publishing the form automatically creates its response sheet, and that sheet is the single system of record from here on.
A TinyGPT step reads each submission and returns structured JSON, not a paragraph: sentiment (positive, neutral, or negative), category (Feature Request, Product Quality, Delivery Experience, Complaint, Praise, or General Feedback), and a canonical theme label. That theme label is the part that matters over time, it's what makes twenty differently-worded complaints about the same thing show up as one trend instead of twenty one-offs.
Rating, comments, sentiment, category, and theme are all written back onto the same response sheet row. There's no second table to reconcile and no export step, reporting and trend analysis read from the same place the customer's answer landed.
Feature requests post to your Product channel in Slack. Product-quality issues go to Product/QA. Complaints and delivery problems go to Support. Each message carries the AI summary and a reference back to the record, so whoever picks it up isn't starting from scratch. Prefer Telegram or Discord? Swap at install.
Clearly positive feedback is flagged testimonial_ready on the row and nothing else happens, no alert, no ping, no channel noise. Your marketing team just pulls the shortlist from the sheet whenever they need real customer words.
Weekly by default, AI reviews recent responses and emails leadership one summary: overall sentiment, the breakdown by category, which issues keep recurring, and what customers are asking you to build. Prefer monthly? Change the schedule node in Studio after install.
Everything it creates is an ordinary form, table and workflow in your workspace. Edit any of it after install. Nothing is locked.
