Most prioritization quietly collapses into “what feels urgent right now.” A good prompt refuses that. It forces a list through impact, effort, decay, and dependency at the same time, and it names what gets dropped — not just what gets done. The 12 templates below do that. Each one paste-and-go, each one ending in a decision, not a vibe.
TL;DR
- Capture your full list first. Prioritizing a half-captured list is wishful thinking, not triage.
- Make the model score on more than one axis: impact, effort, time-decay, and what each task unblocks.
- Always force a “drop / delegate / defer” verdict so nothing sits in limbo.
- For 10-30 tasks an AI 2x2 (impact × effort) is enough; for 30+ ask for a RICE-style score:
(Reach × Impact × Confidence) ÷ Effort. - Paste these into any current model. As of June 2026 a typical weekly list fits comfortably in ChatGPT Plus ($20/mo), Claude Pro ($20/mo, Sonnet 4.6, 1M-token context), or Google AI Pro ($19.99/mo, Gemini 3.1 Pro, 1M context).
Who this is for
Individual contributors drowning in tickets, managers splitting a week between their own work and their team’s, and founders who need to decide what their company will not ship this quarter. The harder your week is to defend out loud, the more these prompts help.
When not to use these prompts
Skip them on a list you haven’t fully captured. AI ranking a partial backlog gives you a confident order over the wrong set of items. Dump everything first — Inbox, Slack flags, the sticky note, the “I should really…” thoughts — then prioritize. Also skip pure-judgment calls (who to hire, whether to pivot); these prompts sort known work, they don’t make strategic bets for you.
Pick a framework before you pick a prompt
The framework matters more than the wording. Match it to list size and what you’re deciding.
| Framework | Best for | The math | List size |
|---|---|---|---|
| Eisenhower (urgent × important) | Quick daily sort | 2x2, no scoring | < 15 |
| Impact × Effort | Defensible weekly triage in ~20 min | 2x2, rough scores | 10-30 |
| RICE | A backlog where you must justify order | (Reach × Impact × Confidence) ÷ Effort | 30+ |
| WSJF | Time-sensitive work competing for one team | Cost of Delay ÷ Job Size, scored on Fibonacci 1,2,3,5,8,13,20 | 30+ |
Eisenhower and impact-effort win on adoption because people actually finish them; RICE and WSJF win on rigor when you need a number to point at in a meeting. Templates 1-3 below cover the first three; ask the model to apply WSJF explicitly if your work has real deadlines.
Which AI model for prioritizing a task list (June 2026)
For weekly or quarterly triage, model choice barely matters — any current frontier model can run a 2x2. Context window only matters once your “list” is a 2,000-line backlog export or a quarter of meeting notes:
- ChatGPT Plus ($20/mo): GPT-5.5 default, in-app context ~320 pages — fine for a week’s tasks, tight for a raw backlog dump. The full 1M-token window needs the $200 Pro tier.
- Claude Pro ($20/mo, $17 annual): Sonnet 4.6 with a 1M-token context — best when you paste a whole project’s notes and want the structure preserved. Bundles Claude Code and Cowork.
- Google AI Pro ($19.99/mo): Gemini 3.1 Pro, 1M context, ties into Google Workspace if your tasks live in Docs and Calendar.
Practical rule: short list, use whatever you already pay for. Huge backlog or a quarter of notes, reach for Claude Pro or Google AI Pro for the 1M-token window.
12 copy-ready prompt templates
Swap each [bracketed] placeholder for your real data before sending. The model can only sort the work it can see, so paste the actual list, not a description of it.
1. Impact × effort matrix
Tasks: [paste your task list]. Score each on IMPACT (1-5) and EFFORT (S / M / L).
Output a 2x2: high impact / low effort (do first), high / high (plan and schedule),
low / low (drop), low / high (do not do). Add a one-sentence rationale per item.Swap: [paste your task list]
2. Decay-aware priority
For each task, classify time-decay: (a) Compounds if delayed (gets more valuable or
costly to delay), (b) Stable, (c) Decays (less valuable later). Then re-rank:
do-soon = Compounding, defer = Stable, drop-or-do-now = Decaying. Explain any task
where decay flips its priority versus a plain impact sort.3. Dependency graph
Tasks: [paste your task list]. Map dependencies: which tasks unblock others?
Output: (a) the critical path, (b) tasks that block 2+ others (do these first),
(c) leaf tasks with nothing depending on them (do last). Flag any circular blocks.Swap: [paste your task list]
4. Drop / delegate / defer
For each task decide exactly one: DO (this week), DELEGATE (name who + a one-line brief),
DEFER (with a calendar date), or DROP (and one line on how to tell the requester).
Leave nothing in limbo. List every DROP at the top so I can sanity-check it.5. Energy / cognitive load
My energy pattern: [describe your day, e.g. sharp 8-11am, slump after lunch, social late].
Group tasks: (a) deep-work, (b) admin, (c) social/meetings. Re-allocate each task to the
slot that matches its cognitive load. Do not schedule deep work between two meetings.Swap: [describe your day]
6. Quarterly priorities (3 things)
For this quarter, identify the 3 priorities that move the most. For each give: the outcome,
one success metric, and what we will NOT do in service of it. If you cannot get the list down
to 3, tell me which ones to cut and why — too many priorities means none.7. Weekly priority memo
Write a 5-bullet "this week's priorities" memo: (1) top priority + why, (2) two secondary
items, (3) what I am saying no to and to whom, (4) decisions I need from others.
Keep it under 150 words. Plain, confident tone.8. Saying-no template
I need to decline `[request]` from `[requester]`. Write 3 options: (a) a clean no with an
alternative, (b) defer with a specific date, (c) yes-if (negotiate scope down).
Each under 50 words, warm but firm, no over-apologizing.Swap: [request], [requester]
9. Backlog grooming
My backlog: [paste items]. Sort into: (a) actionable within 1 week, (b) within 1 month,
(c) more than 1 month or unclear (move to "someday"), (d) drop entirely (no longer relevant).
Output by bucket. For anything you drop, give a one-line reason.Swap: [paste items]
10. Maker vs. manager day plan
I have maker days (deep work) and manager days (meetings). Tasks: [paste list].
Reflow them: which fit a maker day, which fit a manager day? Batch shallow work together.
Never schedule deep work in a 30-minute gap between two meetings.Swap: [paste list]
11. “Smallest visible win” filter
For each big project on my list, identify the smallest visible win I can ship this week —
a tangible artifact (memo, prototype, ticket, demo) that proves momentum to someone else.
Do not propose hidden or prep-only work. One concrete artifact per project.12. Weekly retro
Last week I planned `[planned items]` and completed `[completed items]`. Audit:
(1) why did I miss what I missed (capacity, interruptions, or bad estimate?),
(2) what surprised me (the urgent things I did not plan), (3) one concrete adjustment
for this week. Be specific, not encouraging.Swap: [planned items], [completed items]
Common mistakes
- Prioritizing a partial list. A confident order over the wrong items is worse than no order.
- One axis only. Urgency alone buries the high-impact, non-urgent work that actually compounds.
- No “drop” verdict. If a prompt can only re-rank, your list never shrinks — force drop/delegate/defer.
- Letting AI invent numbers. It will produce plausible reach or revenue figures. Paste real data or leave the cell blank.
- No word limit on memos. Executives skim; cap at 150 words or three bullets.
- Re-prioritizing daily. Thrash kills momentum. Set a triage cadence (weekly) and hold it.
How to push results further
- Tell the model the seniority of the reader so it pitches the rationale right.
- Cap output length: one tactical page, or three executive bullets plus a link.
- Lead every memo with the decision needed, not the background.
- Ask for the drops first — that is where the model is most useful and most checkable.
- Re-run template 12 (retro) against template 6 or 7 from last week to close the loop on your own estimates.
FAQ
- Which AI model is best for prioritizing a long task list? Any current model handles a weekly list. For a huge backlog or a quarter of notes, use a 1M-token context model — Claude Pro (Sonnet 4.6, $20/mo) or Google AI Pro (Gemini 3.1 Pro, $19.99/mo) as of June 2026.
- Can AI replace a manager’s judgment? It drafts, scores, and structures well; the strategic calls (what matters, who to trust) stay human. Treat its order as a strong first draft.
- How often should I re-prioritize? Weekly for tasks, quarterly for projects. Re-ranking more often than your work actually changes just creates thrash.
- Should I include risks in the output? Yes. Ask the model to name what could go wrong per priority. Pretending risk is zero erodes trust the first time something slips.
- How do I stop the model inventing numbers? Never ask it to generate reach, revenue, or effort estimates from nothing. Paste real figures, or have it output the scoring grid with blanks for you to fill.
- Does context window matter for this? Only if you paste a lot. A short list fits anywhere; a 2,000-line backlog export or months of notes needs the 1M-token tiers above.