A quarterly plan with 14 priorities ships zero. Google runs four to six OKRs per quarter, with only one or two objectives at the company level — and it still calls roughly half of them aspirational stretch goals where 70% is a win (see Google re:Work: Set goals with OKRs). The prompts below force the same discipline: 3-5 priorities, an explicit list of what gets cut, capacity math that does not pretend everything fits, and a committed-vs-bet ranking so you stop labeling every line item “must-deliver.”
TL;DR
- Use these 12 templates to draft, pressure-test, and communicate a quarterly plan — not to generate one from nothing.
- Each template targets one job: outcome framing, capacity math, trade-off statements, bet ranking, the kickoff memo, mid-quarter recalibration, and the retro.
- AI drafts and stress-tests; you supply real headcount, real metrics, and the judgment call. AI invents plausible numbers if you let it — paste your own.
- Paste these into ChatGPT (GPT-5.5), Claude (Sonnet 4.6 / Opus 4.7), or Gemini 3.1 Pro. For a recurring plan, a Claude Project or Gemini Gem keeps your team’s context loaded across the quarter.
Who this is for
Function leads and PMs running quarterly planning, founders setting OKRs, and engineering managers committing to a roadmap. If you have a planning template your org already uses, paste it as an example so output matches your house format.
Pick a model and a workspace (June 2026)
The prompts are model-agnostic, but where you run them changes how much context you can load:
| Surface | Model | Why for planning |
|---|---|---|
| ChatGPT Plus ($20) | GPT-5.5 (Thinking) | Strong reasoning; in-app context ~320 pages, so trim attachments. Full 1M-token in-app context is only on the $200 Pro tier. |
| Claude Pro ($20) | Sonnet 4.6 / Opus 4.7 | 1M-token context standard; Claude Projects hold your strategy doc + last quarter’s plan and reread them every chat. |
| Google AI Pro ($19.99) | Gemini 3.1 Pro | 1M context; Gems pull org context straight from Google Drive without re-uploading. |
For a one-off plan, any chat works. For recurring quarterly cadence, load your strategy memo and previous plan into a Claude Project or Gemini Gem so you stop re-pasting context.
Prompt anatomy
Every planning prompt should carry six elements:
- Role: who AI plays — chief of staff, manager, or analyst.
- Context: team, org, scope, and the real data attached.
- Goal: one deliverable — plan, memo, talking points, or doc.
- Constraints: word count, must-include fields, audience seniority.
- Tone: confident, neutral, factual.
- Examples: 1-2 prior samples to anchor the format.
12 copy-ready prompt templates
Swap the backtick [placeholders] for your real values before running. Replace every [metric] with an actual number — never let the model fill those in.
1. Outcome-first priorities
Quarter: `[quarter]`. Generate 3-5 priorities. Each: (a) outcome (not output — what changes for users / business), (b) success metric, (c) why this not something else, (d) what we explicitly won't do.Swap: [quarter]
2. Capacity math
Team has `[n]` people, `[weeks]` weeks effective. Map each priority to estimated person-weeks. Sum. If sum > capacity, list which priorities to cut. Don't pretend everything fits.Swap: [n], [weeks]
3. Outcome vs output check
Audit my priorities for output-thinking ("ship feature X") vs outcome ("X% lift in retention"). Rewrite each as outcome. If output is the goal, name the outcome it serves.4. Trade-off statement
For each priority, write a 1-sentence trade-off statement: "By doing X, we accept slower progress on Y." If you can't name a trade-off, the priority is probably untested.5. Bet ranking
Some priorities are commitments, some are bets. Classify each as: (a) Commitment (deliver at 100%), (b) Bet (we'll learn within the quarter whether to continue), (c) Stretch (70% counts as a win). Don't mark everything a commitment.This mirrors Google’s committed-vs-aspirational split: commitments you deliver fully, aspirational goals you score around 70%.
6. Quarter kickoff memo
Write the team kickoff memo: (1) Our 3-5 priorities, (2) What we won't do, (3) How we'll measure success, (4) Risks we're watching, (5) First milestones. ≤ 500 words.7. Cross-team dependency map
For each priority, list dependencies on other teams: (a) What we need, (b) When, (c) Who owns. Flag any dependency without an "owner" name.8. Mid-quarter check-in
Week 6 of quarter. Audit: (1) Which priorities are on / off track, (2) Should we cut any priorities now to focus, (3) Where the team is over-committed. Output a recalibration memo.9. Quarter retro
End of quarter. Retro: (1) Priorities shipped, (2) Priorities not shipped (and why), (3) What we learned that surprised us, (4) Process changes for next quarter. Output ≤ 300 words.10. “What we won’t learn”
For this quarter, name 3 things we won't learn (questions we won't answer). Naming them surfaces gaps. Decide: which can wait, which need a small bet next quarter.11. Leadership pitch
Pitch the quarterly plan to leadership in 5 sentences: (1) Theme, (2) 3 priorities + outcomes, (3) Trade-off accepted, (4) What we need from them, (5) Risks we're watching.12. Planning hygiene audit
Audit the plan: (1) > 5 priorities? Trim. (2) Any "improve X" with no metric? Make it concrete. (3) Capacity math actually done? (4) Trade-offs explicit? Output fix list.A worked sequence
Most teams chain these rather than run one in isolation:
- Draft with #1, attaching last quarter’s results.
- Reality-check with #2 (capacity) and #3 (outcome vs output).
- Sharpen with #4 (trade-offs) and #5 (bet ranking).
- Communicate with #6 (kickoff memo) and #11 (leadership pitch).
- Track with #8 at week 6, then close with #9.
Common mistakes
- No specific context, so the output is generic.
- Letting the model fill in metrics — it invents plausible numbers. Paste real data.
- Vague audience, so the draft over- or undershoots seniority.
- No word limit, so readers do not finish.
- Marking every priority a “commitment,” which defeats the ranking.
- No “decision needed” framing.
How to push results further
- Specify audience level and cap length: 1 page tactical, 3 bullets executive.
- Lead with the ask or the decision needed.
- Attach the source data instead of describing it.
- Read the memo aloud before sending.
- Save your best outputs as examples and reuse the format, refreshing the content each quarter.
FAQ
- How many priorities should a quarter have?: Three to five at the team level; one or two objectives at the company level. Google runs four to six OKRs per quarter.
- How long should the plan doc be?: Tactical version 1 page; executive version 3 bullets plus a link.
- Can AI replace the analyst or manager?: It drafts and templates well; the judgment, headcount truth, and the “no” are yours.
- Should I include risks?: Always. Pretending no risk exists erodes trust the first time one lands.
- How do I fact-check the numbers?: Attach your sources and have a peer review every metric. Do not ship a number the model produced.
- Can AI generate the data?: No. It fabricates plausible figures. Connect a real spreadsheet or paste the actual numbers.