Multi-channel feedback to Jira pipeline with AI analysis & Notion reporting

Description

This workflow turns scattered user feedback into a structured product backlog pipeline.

• It collects feedback from three channels (Telegram bot, Google Form/Sheets, and Gmail), normalizes it, and sends it to an AI model that:

• Classifies the feedback (bug, feature request, question, etc.)

• Extracts sentiment and pain level

• Estimates business impact and implementation effort

• Generates a short summary

• Then a custom RICE-style priority score is computed, a Jira ticket is created automatically, a Notion page is generated for documentation, and a monthly product report is sent by email to stakeholders.

It helps product & support teams move from “random feedback in multiple tools” to a repeatable, data-driven product intake process with zero manual triage.

Context

In most teams, feedback is:

• spread across emails, forms, and chat messages

• manually copy–pasted into Jira (when someone remembers)

• hard to prioritize objectively

• nearly impossible to review at the end of the month

This workflow solves that by:

• Centralizing feedback from Telegram, Google Forms/Sheets, and Gmail

• Automatically normalizing all inputs into the same JSON structure

• Using AI to categorize, tag, summarize, and score each request

• Calculating a RICE-based priority adapted to your tiers (free / pro / enterprise)

• Creating a Jira issue with all the context and acceptance criteria

• Generating a Notion page for each feedback+ticket pair

• Sending a monthly “Product Intelligence Report” by email with insights & recommendations

The result: less manual work, better prioritization, and a clear story of what users are asking for.

Target Users

This template is designed for:

• Product Managers and Product Owners

• SaaS teams with multiple feedback channels

• Support / CS teams that need a structured escalation path

• Project Managers who want objective, data-driven prioritization

• Any team that wants “feedback → backlog” automation without building a custom platform

Technical Requirements

You’ll need:

• Google Sheets credential

• Gmail credential

• Telegram Bot + Chat ID

• Google Form connected to a Google Sheet

• Jira credential (Jira Cloud)

• Notion credential

• OpenAI/ Anthropic credential for the AI analysis node

• An existing Jira project where tickets will be created

• A Notion database or parent page where feedback pages will be stored

Workflow Steps

The workflow is organized into four main sections:

1. Triggers (Multi-channel Intake)

• Telegram Trigger – Listens for new messages sent to your bot

• Google Form / Sheet Trigger – Listens for new form responses / rows

• Gmail Trigger – Listens for new emails matching your filter (e.g. [Feedback] in subject)

• All three paths send their payloads into a “Data Normalizer” node that outputs a unified structure:

1. Request Treated and Enriched (AI Analysis)

• Instant Reply (Telegram only) – Sends a quick “Thanks, we’re analysing your feedback” message

• User Enrichment – Enriches user tier based on mapping

• Message a Model (AI)

  • classifies the feedback

  • extracts tags

  • scores sentiment, pain, business impact, effort

  • generates a short summary & acceptance criteria

• JSON Parse / Merge – Merges AI output back into the original feedback object

1. Priority Calculation & Jira Ticket Creation

• Priority Calculator applies a RICE-style formula using:

  • pain level

  • business impact

  • implementation effort

  • user tier weight

  • assigns internal priority: P0 / P1 / P2 / P3

  • maps to Jira priority: Highest / High / Medium / Low

• Create Jira Issue – Creates a ticket with:

  • summary from AI

  • description including raw feedback, AI analysis, and RICE breakdown

  • labels based on tags

  • priority based on the calculator

• Post-processing – Prepares a clean payload for notifications & logging

• IF (Source = Telegram) – Sends a rich Telegram message back to the user with:

  • Jira key + URL

  • category, priority, RICE score, tags, and estimated handling time

  • Append to Google Sheet (Analytics Log) – Logs each feedback with:

  • source, user, category, sentiment, RICE score, priority, Jira key, Jira URL

• Create Notion Page – Creates a documentation page linking:

  • the feedback

  • the Jira ticket

  • AI analysis

  • acceptance criteria

1. Monthly Reporting (Product Intelligence Report)

• Monthly Trigger – Runs once a month

• Query Google Sheet – Fetches all feedback logs for the previous month

• Aggregate Monthly Stats – Computes:

  • feedback volume

  • breakdown by category / sentiment / source / tier / priority

  • average RICE, pain, and impact

  • top P0/P1 issues and top feature requests

  • Message a Model (AI) – Generates a written “Product Intelligence Report” with:

  • executive summary

  • key insights & trends

  • top pain points

  • strategic recommendations

• Parse Response: Extracts structured insights + short summary

• Create Notion Report Page with:

  • metrics, charts-ready tables, insights, and recommendations

  • Append Monthly Log to Google Sheet – Stores high-level stats for historical tracking

• Send Email with a formatted HTML report to stakeholders with:

  • key metrics

  • top issues

  • recommendations

  • link to the full Notion report

Key Features

• Multi-channel intake: Telegram + Google Forms/Sheets + Gmail

• AI-powered triage: automatic category, sentiment, tags, and summary

• RICE-style priority scoring with tier weighting

• Automatic Jira ticket creation with full context

• Notion documentation for each feedback and for monthly reports

• Google Sheets analytics log for exploration and dashboards

• Monthly “Product Intelligence Report” sent automatically by email

• Designed to be adaptable: you can plug in your own labels, tiers, and scoring rules

Expected Output

When the workflow is running, you can expect:

• A Jira issue created automatically for each relevant feedback

• A confirmation email

• A Telegram confirmation message when the feedback comes from Telegram

• A Google Sheet filled with normalized feedback and scoring data

• A Notion page per feedback/ticket with AI analysis and acceptance criteria Every month:

• a Notion “Monthly Product Intelligence Report” page

• a summary email with key metrics and insights for your stakeholders

How it works

• Trigger – Listens to Telegram / Google Forms / Gmail

• Normalize – Converts all inputs to a unified feedback format

• Enrich with AI – Category, sentiment, pain, impact, effort, tags, summary

• Score – Computes RICE-style priority and maps to Jira priority

• Create Ticket – Opens a Jira issue + Notion page + logs to Google Sheets

• Notify – Sends Telegram confirmation (if source is Telegram)

• Report – Once a month, aggregates everything and sends a Product Intelligence Report

Tutorial Video

Tutorial video: Watch the Youtube Tutorial video

About me

I’m Yassin a Project & Product Manager Scaling tech products with data-driven project management. 📬 Feel free to connect with me on Linkedin