Loading the exercise
Loading the exercise
Analyze the sentiment and topic of feedback, log the results, then trigger safe, tested reactions.
In this exercise you build a more complex automation with AI using Make.com. Customers of a fictional business send feedback through a short Google Form; Make automatically detects the new feedback, uses AI to determine its sentiment (positive / neutral / negative), identifies its topic (Termék, Ügyfélszolgálat, Szállítás, Egyéb), writes the results back into the Google Sheet, and then decides based on the sentiment: negative feedback triggers a Slack alert, positive feedback sends an automatic thank-you email, and neutral feedback triggers no further action. Here an AI decision has a real consequence — that is why testing, logging and human responsibility are central.
In this exercise you will learn: how a multi-module Make Scenario is built; how Google Sheets can act as a trigger; how to pass data between modules; how AI performs automatic sentiment analysis; how AI classifies text feedback; how an existing Google Sheet row can be updated automatically; how the Router works; how Filters work; how the same data can trigger different events; how to send an automatic Slack alert; how to send an automatic Gmail message; how to test an AI-based automation with multiple inputs; and why it matters to verify an AI-made decision.
Make's current free plan provides 1000 credits per month; scheduled Scenarios can run with a minimum 15-minute interval. Plan conditions can change, so check Make's current pricing before you start. The Watch New Rows and Update a Row modules used in this exercise are available in Make's Google Sheets integration.
Important: The Make AI Toolkit currently also works on the free plan with Make's own AI Provider, so you do not need a separate OpenAI API key or a paid ChatGPT API account for this practice. The Toolkit also provides dedicated Analyze Sentiment and Categorize Text modules. The credit usage of AI modules can depend on the chosen model and the length of the processed text.
Warning: Important: during the lesson always use the Run once test run, and do not leave the Scenario active at the end of the exercise.
Tip: The motto of this exercise: COLLECT. ANALYZE. ROUTE. VERIFY. — the four stages of the flow: 01 FORM (Google Forms) → 02 AI (sentiment + topic) → 03 LOG (row update) → 04 REACTION (Slack + Gmail).
Free preview
The first step is open to everyone. The remaining 21 steps, progress tracking and feedback open up with a free account.