Productivity Mastery path
Free starter lessonChatGPT16 min

Find your biggest time drain with ChatGPT

By the end you will have a one-page time audit, built from your real calendar, that ranks where your week actually goes and names the single tool to deploy first.

You cannot fix a time problem you have not measured, and "I feel busy" is not a diagnosis. Ranking your week by real hours tells you whether meetings, note-taking, admin, or email is the actual drain, so the rest of this path attacks the biggest one instead of the most annoying one.

Ten minutes of data beats a month of buying tools you never use.

This is the free first module of urfired.ai's Productivity Mastery path. No signup, no card. The steps note what ChatGPT access you need.

The workflow

8 steps, start to finish

Pull one real week as data

Fastest path: open last week in your calendar and paste it block by block into a note, one line per event with day, start time, end time, and a short label. If you prefer a file, Google Calendar can export from Settings, then Import and export, then Export, but know what you get: a zip you must unzip, and each .ics inside holds your entire calendar history, not just last week, which is fine because you will constrain the window when you parse it.

Classic Outlook can export a CSV through File, then Open and Export, then Import/Export, choosing Export to a file and CSV format. The new Outlook has no File menu, so if that is your client, use the paste method.

Open ChatGPT and give this work a home

Go to chatgpt.com and sign in; the free tier handles file upload and data analysis, with tighter message and context limits than paid plans. If Projects appears in your left sidebar, create one named Time Audit so the chat, file, and follow-ups stay together and you can rerun this next quarter.

If your plan does not show Projects, a regular chat works fine; you will just save the finished audit somewhere durable at the end.

Upload the week and ask for a clean table

Start a new chat (inside the project if you made one) and attach your file or paste your week. If the uploader rejects a .ics file, rename it to .txt, since it is plain text, or open it and paste the contents.

Then prompt: "Only include events between Monday [date] and Sunday [date]. Parse them into a table with columns Day, Start, End, Duration in hours, and Title.

Do not summarize yet, just show the parsed rows so I can confirm they are right." The date window matters because a Google export contains your whole history; skim the table and confirm it caught last week's real events and nothing outside it.

Have it categorize and rank the hours

Next prompt: "Bucket every event into categories: external meetings, internal meetings, note-taking and recaps, email and admin, and focused work. Total the hours per category for the week, rank them highest to lowest, and show each as hours and as a percentage of my logged time." This ranking is the spine of your one-pager.

Clean the obvious errors

Raw exports lie. A multi-day trip can show as one 72-hour block, and blocked lunches inflate the totals.

Tell it what to fix, for example: "Remove my daily lunch block, and treat the Out of Town event as travel, not a meeting." If titles are vague, ask it to re-bucket using the attendee count so a 6-person call is not filed as focused work. Rerun the totals.

Reconcile the total against a number you know

Before you trust the ranking, sanity-check it. Ask: "What is the total logged across all categories?" If you work roughly 45 hours and it reports 78, something is still double-counted.

Push back until the total matches your real week within an hour or two. This is the human check the rest of the path depends on.

Name the single tool to deploy first

Now connect the top drain to one tool. Prompt: "My biggest bucket is X.

From this stack, which one tool should I deploy first and why: Granola or Fireflies for meeting capture, Wispr Flow for voice drafting, Motion for scheduling, Superhuman for inbox, Notion AI for status writeups." Read its reasoning, then make the call yourself. The biggest bucket picks the tool, not the shiniest demo.

Have it write the one-page audit

Final prompt: "Write this as a one-page audit: total hours logged, a ranked table of hours and percent per category, the single biggest drain in one sentence, the one tool to deploy first, and the one habit I should stop. Keep it to one page." Read it, fix anything that misreads your week, and save it: in the project if you made one, or pasted into a note you keep.

That saved page is your deliverable and your baseline to measure against later.

Do this now

Capture your own last 7 days, paste the week block by block into a note, or export it from your calendar if you prefer a file, then feed it to a new ChatGPT chat (inside a Time Audit project if your plan shows Projects), constrain the parse to that Monday-to-Sunday window, and drive it through parse, categorize, clean, and reconcile until you have a one-page audit that ranks your real hours by category and names the one tool from the path stack to deploy first. Use your actual week, not a sample.

What you keep

A one-page time audit, saved in your Time Audit project or somewhere you can rerun it next quarter, that ranks where your week actually goes by hours and percent per category and names the single tool to deploy first against your biggest drain.

What good looks like
  • Every event in your week lands in a named category, and the category totals reconcile to your real weekly hours within an hour or two.
  • The audit ranks categories high to low with an actual number next to each, and the top drain is unambiguous.
  • It names exactly one tool from the stack (Granola, Fireflies, Wispr Flow, Motion, Superhuman, or Notion AI) tied to that top drain, with a one-line reason.
  • You corrected at least one miscategorized block, so the numbers reflect your week and not the raw export.
Common traps
Describing your week in prose ("I'm slammed with meetings") instead of feeding it real calendar data, so the ranking is a guess dressed up as analysis.
Give it your actual week, block by block or as a real calendar export; make it rank facts, not your impression of the week.
Trusting the category totals as printed, when travel events, all-day blocks, and lunch holds have inflated them.
Reconcile the total against a number you already know, like your real working hours, and make it fix the miscounts before you believe the ranking.
Picking the tool you already wanted instead of the one the data points to.
Let the largest bucket choose the tool; if note-taking is the top drain, that is a Granola or Fireflies problem, not an inbox one.
Saved on this device
Next in this path

Module 02: Capture client calls bot-free with Granola (Granola, 18 min)

Free gets you this starter lesson today. Pro unlocks the full 12-module path when checkout opens, no card and no charge until then.

New here? The free 2-minute assessment reads your role and gaps, then matches you to the right path. Get your score