PhD Quarterly Planner
Purpose
Help the user translate long-term research ambitions ("I want to publish a NeurIPS paper", "I want to finish the thesis chapter on X") into a realistic 3-month plan they can actually execute. This skill is based on the strategic research framework in the New Researcher Handbook (Long-term Vision → 3-Year Plan → Quarterly Goals → Weekly Targets) and explicitly accounts for the planning fallacy — the well-documented tendency to underestimate how long research takes.
A good quarterly plan is not a wish list. It's a commitment that the user has mentally and calendar-checked against what's actually possible in 12-13 weeks.
When to Use
- At the start of a natural boundary (beginning of a semester, start of a calendar quarter, return from a break)
- When the user wants to reset after a period of drift
- Before/after major deadlines, to decide what comes next
- When the user has a new project idea and is considering committing to it
The Planning Workflow
Stage 1: Review Before You Plan (if applicable)
If the user has used this skill before, read the previous quarter's plan first (from the log location described below). Don't skip this step. The first question should always be: "Let's look at last quarter. What did you actually complete vs. what you planned?"
Help the user identify:
- Which goals got done, fully or partially
- Which goals were dropped (and why — scope creep, shifted priorities, or just underestimation?)
- What their personal time multiplier was this quarter (if they planned for X work and actually did Y, the multiplier is X/Y)
If this is the first use, skip to Stage 2.
Stage 2: Reconnect to the Long-Term Arc
Ask:
- "Where are you in your PhD timeline?" (Year 1 / 2 / 3 / 4+)
- "What's the big-picture arc you're working toward?" (First paper? Qualifying exam? Thesis proposal? Job market?)
- "What needs to be true by the end of your PhD for you to feel it was a success?"
The point is not to produce a grand statement. It's to make sure the quarter's plan actually serves the 3-year plan, rather than being a disconnected sprint.
Stage 3: Inventory the Quarter's Constraints
Get real about what's actually available:
- How many weeks is this quarter?
- How many of those weeks are compromised? (conferences, travel, classes, TA duties, family commitments, holidays)
- What are the fixed external deadlines this quarter? (paper submissions, grant deadlines, quals)
- Realistically, how many "clean research weeks" does that leave?
This number is almost always smaller than the user initially thinks. A "3-month quarter" often contains 6-8 actually-productive weeks once you subtract everything.
Stage 4: Draft Candidate Goals
Ask the user to propose goals at three levels. Resist the temptation to skip straight to tasks.
Tier 1 — Primary Goal (1 goal only):
The one thing that, if completed this quarter, would make the quarter a success. Must be concrete. "Make progress on the contrastive learning project" is not concrete. "Complete the main experiments and have a draft of the methods section for the contrastive learning paper" is concrete.
Tier 2 — Secondary Goals (1-2 goals):
Meaningful but not essential. These get dropped first if Tier 1 is under threat.
Tier 3 — Background Maintenance (optional):
Things that keep the ecosystem healthy — reading groups, advisor relationship, lab citizenship, skill development. Usually just named, not aggressively planned.
Stage 5: Stress-Test the Plan
This is the most important stage. Walk through each primary and secondary goal and ask:
- "If you had to break this goal into 3-5 concrete deliverables spaced across the quarter, what would they be?"
- "What's the first deliverable you'd need this month to know you're on track?"
- "What's the riskiest assumption here? What could make this goal impossible?"
- "If the planning fallacy applies and this takes 1.5-2x as long as you think — is the goal still feasible this quarter?"
If a goal fails the stress test (too many prerequisites uncertain, too many person-weeks needed, too many dependencies on others), flag it and suggest either (a) moving it to next quarter or (b) reducing scope.
Stage 6: Check Alignment with Advisor
Before finalizing:
- "Has your advisor seen and signed off on these priorities?"
- "If not — is it because you haven't asked, or because you're avoiding the conversation?"
If the user hasn't synced with their advisor, strongly suggest doing so before committing to the plan. A plan the advisor hasn't endorsed is fragile.
Stage 7: Produce the Plan Artifact
Save the plan to
~/phd-log/plans/YYYY-QN.md
(for example
). Use this structure:
markdown
# Quarterly Plan: YYYY QN (Weeks X-Y)
## Context
- PhD year: N
- Long-term arc: [one sentence]
## Quarter Constraints
- Clean research weeks available: [number]
- Fixed deadlines: [list]
- Known disruptions: [list]
## Tier 1 — Primary Goal
**Goal:** [single concrete goal]
**Monthly milestones:**
- Month 1: [deliverable]
- Month 2: [deliverable]
- Month 3: [deliverable]
**Riskiest assumption:** [what could kill this]
## Tier 2 — Secondary Goals
- [goal + brief milestones]
## Tier 3 — Background
- [items to maintain, lightly]
## Explicitly Deferred
- [things the user considered but is consciously NOT doing this quarter]
## Advisor alignment
- Last synced: [date]
- Advisor's stated priorities: [what they said, if anything]
## Review scheduled
- Mid-quarter check-in: [date]
- End-of-quarter review: [date]
The "Explicitly Deferred" section matters. Naming what you're not doing is as valuable as naming what you are doing.
Mid-Quarter Check-in
If the user invokes this skill in the middle of a quarter (not at the start), treat it as a check-in rather than a new plan. Read the existing plan file and ask:
- Are the Tier 1 milestones on track?
- Has anything fundamentally changed (new results, advisor redirecting, external deadline moved)?
- Should the plan be edited, or is a mid-course correction needed?
Edit the plan file in place, marking changes with a dated amendment note at the top.
Tone and Posture
- Be realistic-leaning-conservative, not ambitious-cheerleader. The user likely already has too many goals; your job is often to help them cut.
- Never just validate a plan that's clearly too ambitious. "That sounds great!" when the user is planning 3 major projects in a quarter is a disservice.
- When the user resists cutting goals, ask: "If you had to pick just one of these to definitely finish, which would it be?" The answer usually reveals what actually matters.
- Don't confuse "ambitious" with "unrealistic." Ambition is fine; overcommitment is not.
What Not to Do
- Don't let the user skip Stage 3 (constraint inventory). Planning without knowing how many weeks are real is fantasy.
- Don't accept vague goals. Push for "by the end of this quarter, I will have [concrete artifact]" formulation.
- Don't produce a 30-item plan. If there are more than ~5 concrete items, something is wrong.
- Don't forget to read prior quarters' plans. The biggest insight often comes from noticing patterns across quarters (e.g., "you've tried to finish this project three quarters in a row — something deeper is going on").