Which AI works best for 168-hour audit analysis?

Claude and ChatGPT both handle time-log analysis well. Claude tends to produce more nuanced pattern analysis and is better at surfacing non-obvious connections between your energy ratings and activity categories. ChatGPT is strong on structured output like tables and calculations. Either will work for all five of these prompts — the prompt structure matters more than the specific tool.

Do I need to format my log specially before using these prompts?

No special formatting is required, but clean, consistent input produces better output. A simple list with one entry per line — day, time range, activity, optional notes — is enough. The AI will ask for clarification on ambiguous entries rather than silently miscategorizing them if you flag that preference in the prompt.

5 AI Prompts for Your 168-Hour Audit

You’ve tracked your week. Here are the five prompts that turn a raw time log into decisions.

Copy them, fill in the brackets, and use them in order. Each one builds on the previous.

Prompt 1: The Categorization and Totals

Use this immediately after you finish tracking.

Here is my 168-hour time log for the past week. Each entry shows: 
day, time range, activity, and optionally an energy rating (1–3) or notes.

[paste your log]

Please:
1. Categorize each entry into: Deep Work, Shallow Work, Meetings, Commute, 
   Sleep, Exercise, Meals, Family/Social, Personal Care, Household Tasks, 
   Leisure (Active), Leisure (Passive), Unaccounted
2. Flag any entry where you're uncertain about the category — ask me 
   rather than assume
3. Calculate total hours per category for the full week
4. Calculate weekday totals vs. weekend totals for each category
5. Rank categories from most to least time consumed
6. Tell me my total deep work hours and what percentage that represents 
   of my combined work time (Deep Work + Shallow Work + Meetings)

Why this prompt works: The explicit instruction to flag uncertainty rather than assume prevents silent miscategorization of the entries that matter most — ambiguous work activities and hybrid blocks. The weekday/weekend split often reveals patterns that week-level totals obscure.

Prompt 2: The Priority Alignment Check

Run this once you’ve confirmed the categorization.

My three top priorities right now are:
1. [Priority — be specific about the project or area]
2. [Priority]
3. [Priority]

Based on my categorized time log:
1. How many hours last week went to activities that directly advance each priority?
2. What percentage of my discretionary waking hours (total minus sleep, 
   personal care, and meals) did each priority receive?
3. What is the single biggest gap between a stated priority and actual allocation?
4. If I wanted to add 4 hours per week to [your top priority], which category 
   has the most room to give — and when in the week do those hours occur?

Why this prompt works: Most people run time audits and look at the totals without connecting them to what they said matters most. This prompt does that connection explicitly. The “when in the week” addition converts an abstract reallocation into a concrete scheduling opportunity.

Prompt 3: The Lost Hours Finder

From my categorized log:
1. What are my total Unaccounted hours for the week?
2. What are my total Leisure — Passive hours?
3. What time of day do these two categories tend to occur? 
   (Give me the most common 2-hour windows for each)
4. Are there any recurring patterns — specific times or days where 
   unaccounted or passive leisure consistently appears?
5. If energy ratings are in my log: what is the average energy rating 
   during these windows?
6. How many of these hours occur during windows that could realistically 
   hold focused work or active recovery instead?

Why this prompt works: The “lost hours” are the most reallocatable time in a typical week, and they cluster in predictable patterns. Naming those patterns — “you consistently drift for 30 minutes after every meeting” — is more actionable than a category total.

Prompt 4: The Redesign Model

Use this after you’ve identified what you want to change.

I want to add [X hours] per week of [activity — e.g., deep work on Project X, 
exercise, reading].

Based on my audit data:
- My highest-energy windows are: [list from your analysis]
- My current commitments that can't move: [list fixed meetings, obligations]
- The category most available for reallocation: [from Prompt 3 results]

Please:
1. Suggest the specific time windows that best fit this addition
2. Show me what would need to move or shrink to create those windows
3. Give me a concrete Monday–Friday schedule for Week 1 of this change
4. Tell me the most likely obstacle to maintaining this change 
   based on my log patterns

Why this prompt works: Abstract intentions (“I’ll work out more”) don’t become habits without a concrete schedule. This prompt forces a specific, real-calendar-compatible plan rather than a theoretical one, and the obstacle identification at the end is a miniature pre-mortem.

Prompt 5: The Quarterly Comparison

Use this when you run your second audit three months later.

Here is my Week 1 audit baseline:
[paste summary from Audit 1]

Here is my Week 2 audit data:
[paste your new log or its summary]

Please:
1. Calculate the change in hours per category between the two audits
2. Tell me which category changes were largest (positive and negative)
3. Tell me whether my deep work hours increased, decreased, or held steady
4. Check whether the structural change I committed to after Audit 1 
   ([describe the change you made]) appears in the data — did it hold?
5. Identify any categories that increased unexpectedly
6. Suggest the one most worthwhile change to focus on for the next quarter

Why this prompt works: The longitudinal comparison is where the audit framework earns its real return. A single audit shows you where you are. Comparing audits shows you whether you’re changing — and whether the specific changes you intended to make actually landed in the schedule.

For the full walkthrough of using these prompts with Beyond Time, see The Beyond Time 168-Hour Audit Walkthrough. For the complete step-by-step guide, see How to Do a 168-Hour Audit with AI.

Your action for today: Use Prompt 1 right now if you have your log ready. If you haven’t started tracking yet, save these prompts somewhere accessible — they’ll be waiting when you finish the week.

Frequently Asked Questions

Which AI works best for 168-hour audit analysis?

Claude and ChatGPT both handle time-log analysis well. Claude tends to produce more nuanced pattern analysis and is better at surfacing non-obvious connections between your energy ratings and activity categories. ChatGPT is strong on structured output like tables and calculations. Either will work for all five of these prompts — the prompt structure matters more than the specific tool.
Do I need to format my log specially before using these prompts?

No special formatting is required, but clean, consistent input produces better output. A simple list with one entry per line — day, time range, activity, optional notes — is enough. The AI will ask for clarification on ambiguous entries rather than silently miscategorizing them if you flag that preference in the prompt.