The AI Delegation Triage Prompt: Build Your Q1 AI Action Plan in 5 Minutes

What you'll learn in this article:

• Why "use AI more" fails as a resolution and what replaces it
• The exact prompt structure that converts any task list into a prioritized AI action plan
• Which task types consistently produce the highest ROI when automated
• How to read and act on the triage output without over-committing
• The 90-day automation curve for compounding returns across Q1-Q3

A VP of Operations at a 90-person logistics company made the same decision every year-end for three years: use technology more effectively. The resolution failed every year: not because AI is not useful, but because the resolution had no specificity. Which tasks should AI handle? Which tool? What does the workflow look like?

Without those answers, "use AI more" becomes another browser tab that gets opened and closed without changing anything. The executives who made the most material productivity gains from AI in 2024 did not decide to "use AI more." They identified specific tasks, assigned specific tools, and made deliberate decisions about what to automate and what to keep human.

This prompt produces that analysis in 5 minutes. Time to value: 5 minutes.

The Prompt

Works on GPT-4o, Claude 3.5 Sonnet, or Gemini Advanced. Paste the full prompt and fill in your task list where indicated. The more specific your task descriptions, the more specific the output.

THE DELEGATION TRIAGE PROMPT

"Here is my list of recurring weekly tasks. For each task, classify it on three dimensions: (1) AI automation potential: HIGH / MEDIUM / LOW: and explain why in one sentence. HIGH means AI can handle the majority of this task today with a good prompt or setup. MEDIUM means AI can significantly accelerate this task but a human still needs to direct and review it. LOW means AI can assist only marginally or requires significant human judgment throughout. (2) Estimated weekly time saved if automated: in minutes. Be specific. If the savings depends on setup quality, give a range. (3) Specific tool and prompt pattern I should use: name the exact tool and describe the prompt approach in one or two sentences. Do not recommend tools I would need to pay for separately: I have access to: [list your current tools]. After classifying all tasks, give me a prioritized Q1 action plan. Select the 3 highest-ROI automations to implement first. For each of the 3, provide: (a) the specific setup steps; (b) the estimated one-time setup time; (c) the expected ongoing weekly time savings. My recurring weekly tasks: [PASTE YOUR 10-20 RECURRING TASKS HERE]"

How to Read the Output

The classification table: Ignore LOW items for now. Focus on HIGH items with more than 30 minutes per week first. MEDIUM items are often more interesting: "AI significantly accelerates this task" describes most high-value knowledge work. MEDIUM items require specific setup to realize savings, not just an ad-hoc prompt.

The prioritized action plan: Before acting, verify: is the estimated time savings realistic given how your specific version of this task actually works? AI models estimate conservatively on complex tasks and occasionally overstate savings on simple ones. Recalibrate before committing setup time.

📈 Common Output Patterns

Most executives see these tasks cluster at the top of the HIGH list: weekly reporting and summary writing, research and competitive scanning, first-draft communications. Consistent LOW items: strategic decisions requiring judgment with incomplete information, relationship management conversations, real-time situational awareness tasks.

The Failure Modes

Vague task descriptions produce vague recommendations. "Handle communications" produces generic output. "Draft responses to 15-20 client update requests per week, each requiring a 3-5 sentence response acknowledging their question and updating on project status" produces specific output. Fix: spend 5 minutes adding one sentence of context to each task before running the prompt.

Acting on all 3 simultaneously. Executing three setups in the same week results in all three being half-implemented and none production-ready. Block three separate afternoons, one per workflow, spaced 2 weeks apart.

Not revisiting the classification in Q2. The task list changes. New tools arrive. Models improve. Run the prompt again in April with an updated task list: you will be surprised how much has shifted in 90 days.

The 90-Day Automation Curve

Q1: implement the top 3 HIGH automations. Q2: revisit the top MEDIUM items: after 90 days using the HIGH automations, the MEDIUM ones look different because your baseline has shifted. Q3: run the prompt again with an updated task list. An executive who implements 3 automations per quarter averaging 30 minutes per day in savings recovers roughly 90 hours per quarter by Q4. That is more than two full work weeks: reinvested in high-judgment work that cannot be delegated.

Action Steps Summary

• List your tasks : write out 10-20 recurring weekly tasks with one sentence of context each.
• Run the prompt in ChatGPT Plus, Claude, or Gemini Advanced with your task list filled in.
• Read the classification table : focus on HIGH items with more than 30 minutes savings first.
• Recalibrate time estimates against your real experience before committing to a setup.
• Schedule 3 setup sessions before the end of January: one per priority workflow, each a 60-minute calendar block.