Most AI tracking conversations stay shallow because the prompts are too vague. “How am I doing on my goal?” produces a generic response. A specific, well-structured prompt produces a useful one.
These five prompts cover every stage of an AI tracking practice. Copy them, fill in your details, and use them as-is. They’re designed to work with any AI tool.
Prompt 1: The Weekly Check-In
Use this every week after logging your progress.
I'm doing my weekly goal check-in.
Goal: [your goal in one sentence]
Target: [your 90-day target]
Current week: Week [X] of [Y]
Starting baseline: [where you began]
This week's data:
- Primary outcome metric: [number] (target pace: [number])
- [Process metric 1]: [actual] / [target]
- [Process metric 2]: [actual] / [target]
- [Process metric 3]: [actual] / [target]
Context: [1-3 sentences about the week — what was different, what helped, what got in the way]
Please give me:
1. A one-sentence status assessment (ahead/on track/behind + why)
2. One specific pattern or observation from this week
3. One concrete focus or action for next week
4. One question for me to reflect on before next week's check-inWhy it works: The combination of pace tracking, context, and the reflection question ensures you get something actionable — not just a summary of what you already know.
Prompt 2: The Pattern Analysis
Use this after four or more weeks of tracking. Paste in your last four to six weekly logs.
I'm going to share my last [X] weeks of goal tracking data. I want you to read through all of it before responding.
My goal: [goal]
My 90-day target: [target]
Here are the weekly logs:
[paste each week's data in consistent format]
After reviewing all of this, please tell me:
1. The two or three most significant patterns in the data — not just trends, but correlations between different metrics or context factors
2. What my best weeks had in common (look across all data, not just the outcome metric)
3. What my worst weeks had in common
4. One trend that concerns you if it continues at its current rate
5. The single question you'd most want me to answer to understand my progress better
Don't summarize what I already know. Tell me what the data shows that I might not be seeing myself.Why it works: The instruction “don’t summarize what I already know” is the key phrase. It pushes the AI toward analysis rather than recapping your own data back at you.
Prompt 3: The Falling Behind Recovery
Use this when you’re behind and need to figure out what to do.
I'm behind on my goal and I need help figuring out what to do.
Goal: [goal]
Original target: [target] by [date]
Current status: I'm at [where you are], which means I'm [X% / X units] behind pace
Here's what's happened over the last [X] weeks that explains the gap:
[honest summary — what went wrong, what you deprioritized, what changed in your circumstances]
I need to decide: Is this recoverable? If yes, what specifically needs to change? If not, what does a realistic reset look like?
Please help me think through:
1. Whether the original target is still achievable, and what would need to be true for it to happen
2. If it's not achievable, a revised target that's still meaningful and realistic
3. The single most important behavior change for the next 30 days — not three things, just one
4. What I should stop tracking or stop trying to do to free up capacity for the priority
Be direct. I'd rather know the honest situation than get a pep talk.Why it works: The “be direct” instruction at the end matters. AI tends toward encouragement by default. You’re giving it explicit permission to give you the unvarnished assessment.
Prompt 4: The Monthly Progress Summary
Use this at the end of each month to generate a clean synthesis.
I'm generating my monthly progress summary for [month].
Goal: [goal]
Target: [target] by [date]
Here are my four weekly tracking logs for this month:
[paste all four weeks]
Please generate a monthly summary with these sections:
1. Headline status: one sentence on where I am relative to target
2. Top three wins this month (specific, not generic)
3. Top two or three obstacles or friction points
4. Key metrics trend: for each metric, is it improving, flat, or declining — and by roughly how much?
5. One strategic recommendation for next month based on this data
6. A two-sentence narrative I could share with an accountability partner
Keep it honest and specific. I'm using this to make real decisions, not to feel good about myself.Why it works: The accountability partner narrative forces the AI to write something clear and honest enough to share — which improves the quality of the analysis even if you never share it.
Prompt 5: The Quarterly Goal Audit
Use this every 90 days to assess whether the goal is still the right goal.
I've been tracking [goal] for approximately [X] weeks. It's time for a quarterly audit.
Here's my full performance summary for the period:
[summary of all months — key metrics, trajectory, major events]
I want to do a thorough audit. Please help me think through:
1. Have I made meaningful progress? Be honest about what the data actually shows.
2. Is this goal still the right priority for the next 90 days? What would I be giving up by continuing vs. deprioritizing it?
3. Are my current metrics still measuring the right things, or have I learned something that suggests I should track differently?
4. If I continue: what's the one thing that most needs to change about my approach?
5. If I stop or pivot: what does a clean, intentional ending look like?
6. The hardest question you'd ask me if you were responsible for my success
Don't default to "keep going." Take the case for stopping or pivoting seriously.Why it works: The instruction “don’t default to keep going” is important. Quitting is often the right decision, but AI — like most advisors — tends toward continuation. You’re explicitly asking it to steelman the other path.
These prompts are starting points, not scripts. The best version of each will be adapted to your specific goal, your language, and the kinds of insights that have been most useful in previous conversations.
Your action for today: Copy Prompt 1 (the weekly check-in), fill in your data from this week, and run it. Don’t wait for the “right” week or the “complete” data. Run it now with what you have.
Frequently Asked Questions
-
Do these prompts work with any AI tool?
Yes — these prompts work with ChatGPT, Claude, Gemini, or any other conversational AI. The key is giving the AI enough context about your goal, your target, and your current data. A better prompt with more context will always outperform a shorter prompt with less context, regardless of which tool you're using.
-
How often should I use these prompts?
Prompt 1 (weekly check-in) is designed for weekly use. Prompt 2 (pattern analysis) works best after you have four or more weeks of data. Prompt 3 (falling behind) is for when you need it. Prompt 4 (monthly summary) is monthly. Prompt 5 (quarterly audit) is every 90 days. Together, they cover the full rhythm of an effective AI tracking practice.