Create an AI-Powered Project Tracker

You'll end up with: A live Airtable project tracker that turns rough project ideas into prioritized tasks, next actions, owners, due dates, and a weekly AI review habit.

Overview
25-35 min
Beginner
Free
2 tools
Cost breakdown
AirtableFree
Claude or ChatGPTFree
TotalFree
Common mistake

Building too many fields before the tracker has a weekly habit. Keep the first version to project, owner, status, due date, impact, effort, next action, blocker, and notes. Let AI help prioritize and summarize, but make the final status/date decisions yourself.

Before you start
  • A list of 5-15 active or upcoming projects
  • Access to Airtable
  • Access to Claude or ChatGPT
  • Know your next 1-2 deadlines
  • Decide whether this tracker is personal, client-facing, or team-facing
1

Decide the tracker fields before opening Airtable

Turn a messy project list into a simple schema that is small enough to maintain weekly.

Claude or ChatGPTFreeOpen Claude or ChatGPT
Exact action

Paste your current project list into one chat and ask the AI to normalize it into these fields: Project, Outcome, Status, Owner, Due date, Impact, Effort, Next action, Blocker, Notes. Tell it to use only the statuses Not started, In progress, Blocked, Waiting, and Done.

You have a clean table draft with one row per real project and no more than 10 fields.
The AI invents complex columns like risk matrices, dependencies, or milestones. Re-prompt: Simplify this for a solo operator who will update it once per week.
2

Build the Airtable base from the cleaned table

Create the working tracker and set the minimum useful field types.

AirtableFreeOpen Airtable
Exact action

Create a new Airtable base named AI Project Tracker. Add a Projects table. Create fields for Project, Outcome, Status, Owner, Due date, Impact, Effort, Next action, Blocker, and Notes. Use single select for Status, single select or number for Impact and Effort, date for Due date, and long text for Outcome, Next action, Blocker, and Notes. Paste or import the rows from step 1.

Every project has a status, due date or No date yet, and one concrete next action.
More than 20% of rows have blank Next action. Go back to the AI and ask it to infer the smallest next action from the project outcome.
3

Add an AI priority pass without over-automating

Use AI to rank what deserves attention this week, then copy the result back into Airtable.

Claude or ChatGPTFreeOpen Claude or ChatGPT
Exact action

Export or copy the Airtable rows into the same chat and ask: Rank these projects for this week using Impact, Effort, Due date, Blocker, and Status. Return: Project, Priority 1-5, Why now, Suggested next action. Do not change due dates. Do not mark anything done. Add fields in Airtable for Priority, Why now, and AI suggested next action, then paste the outputs into the matching rows.

The top 3 projects are obvious and each has a short reason tied to urgency, leverage, or blocker removal.
The AI ranks everything as urgent. Re-prompt it to force exactly three Priority 5 projects and explain what gets deferred.
4

Create views for daily use, blocked work, and weekly review

Make the tracker useful without searching or filtering every time.

AirtableFreeOpen Airtable
Exact action

Create three views: 1. This Week, filtered to not Done and sorted by Priority then Due date. 2. Blocked, filtered where Status is Blocked or Blocker is not empty. 3. Weekly Review, grouped by Status and sorted by Priority. Hide Notes in the first two views so the tracker stays scannable.

You can open This Week and know what to work on within 30 seconds.
The view still feels noisy. Hide any field that does not change the next decision.
5

Save a weekly AI review prompt and schedule the habit

Turn the tracker into a recurring operating rhythm.

Airtable plus Claude or ChatGPTFreeOpen Airtable plus Claude or ChatGPT
Exact action

Create an Airtable long-text field or a separate note called Weekly review prompt. Save this prompt: Here are my current project rows. Identify: 1) top 3 projects for this week, 2) blocked projects that need a decision, 3) projects that should be paused or deleted, 4) unclear next actions, and 5) one suggested update per row. Do not invent new projects. Add a recurring calendar reminder for Friday or Monday to copy the current rows into the AI chat and paste the recommendations back into Airtable.

Your tracker has a repeatable weekly update ritual and no project is relying only on memory.
The weekly review creates too many suggestions. Add: Limit recommendations to the five changes with the highest leverage.

All done!

You now have: A live Airtable project tracker that turns rough project ideas into prioritized tasks, next actions, owners, due dates, and a weekly AI review habit.

Explore more guides

Want this workflow built for your business?

Book a free audit