20 of the best prompts for Claude prompts for planning, step by step across 4 stages. Works with ChatGPT, Claude, and Gemini.
20 of the best prompts for Claude prompts for planning, step by step across 4 stages. Works with ChatGPT, Claude, and Gemini.
Published July 14, 2026
Most people try to use AI for Claude Prompts for Planning with a single vague prompt and get generic results. This guide takes a different approach: 4 targeted stages, from Define what you are actually trying to do through Review and learn, each with a prompt that gives the AI exactly the context it needs. Use Claude to build rigorous plans: goal clarification, milestone mapping, weekly scheduling, and adaptive review cycles that turn intentions into executed projects. Works with ChatGPT, Claude, and Gemini.
Claude is exceptionally good at challenging vague goals and surfacing hidden assumptions. These prompts use that strength to get precise about what you are planning before building anything.
Goal interrogation
I want to [STATE YOUR GOAL]. Before we plan anything, interrogate this goal: What am I assuming is true that might not be? Is this outcome or activity? Is this what I actually want or what I think I should want? What problem does achieving this solve? What do I lose if I achieve it? Push back on anything vague or unexamined. I want to be sure I am planning the right thing before I invest effort in the plan.
Success definition
My goal is [GOAL]. Help me define success precisely: the specific observable state that means I have succeeded (not "I feel good about it" but what is actually true in the world), the metrics or evidence I would use to evaluate it, the date by which I would call it a failure if not achieved, and the minimum version of success that is still worth pursuing. Fuzzy goals produce fuzzy plans.
Assumption mapping
I am planning to [PROJECT / GOAL]. Map the assumptions baked into this plan: What must be true for this to work? Which of these assumptions am I most certain about versus most uncertain about? Which assumptions, if wrong, would make the whole plan fail? What is the single most important assumption I should validate before investing significant time or money?
Scope negotiation
I want to [GOAL] and I have [TIME AVAILABLE: HOURS PER WEEK] and [RESOURCES]. What is the right scope for what I can realistically do? Tell me: the version of this goal that fits my constraints (not the ideal version), what I am sacrificing in the reduced scope and whether that changes whether it is worth pursuing, and the minimum viable version that still produces real value. Help me make a real commitment, not an aspirational one.
Competing priorities audit
I want to commit to [GOAL] but I already have: [LIST OTHER COMMITMENTS]. Honestly assess whether I have room: what would have to move or shrink to make space for this, what I am underestimating about the time or energy this will take, and whether I am planning to add this to a full plate or actually create room. A plan I cannot execute is worse than no plan, it just creates guilt.
Claude can reason about sequences, dependencies, and constraints with precision. These prompts use that to build a plan structure that holds up.
Backwards planning
My goal: [GOAL]. My deadline: [DATE]. Work backwards from the deadline: what must be complete one month before, one week before, and one day before for this to succeed? Then identify the critical path, the sequence of steps where a delay in one delays everything, and the parallel work that does not block the critical path. This tells me where to protect time and where I have flexibility.
Dependency mapping
Here are the major steps in my plan: [LIST STEPS]. Map the dependencies: which steps must complete before others can start, which can happen in parallel, which have external dependencies (waiting on others, approvals, deliveries), and which are the highest-risk dependency nodes where a block would cascade. Then tell me the order I should actually work in versus the order I listed these.
Resource plan
My project: [DESCRIBE]. The resources I have: [TIME, MONEY, TOOLS, PEOPLE, SKILLS]. Build the resource allocation: how each resource gets distributed across the phases of the project, where I am likely to be under-resourced, what I can do to extend a constraint (buy more time, hire, learn, use a tool), and what happens to the plan if a resource is cut by [30%]. Tell me now rather than mid-project.
Pre-mortem
My plan for [GOAL]: [DESCRIBE PLAN]. Run a pre-mortem: imagine we are six months from now and this plan has completely failed. What are the three most likely causes of that failure? For each: the early warning sign I should watch for, what I can build into the plan now to reduce the risk, and whether any failure mode is severe enough to reconsider the plan before starting.
First week in detail
My plan for [GOAL] starts now. Design the first week in enough detail that I can execute without deciding anything: the specific tasks each day, in order, with realistic time estimates, the single most important outcome this week needs to produce for the project to be on track, and the one thing I should not do this week even though it might feel productive. Momentum in week one determines whether the plan takes hold.
Claude can serve as a thinking partner for the problems that come up during execution. These prompts handle the decisions, stuck points, and drift that happen between plan and completion.
Decision under uncertainty
I need to make a decision about [DECISION] and I do not have complete information. My options are: [LIST OPTIONS]. Help me decide: what information would change my decision (and can I get it quickly), what the cost of delay is versus the cost of a wrong decision, which option is most reversible if I learn I was wrong, and what I would tell a friend in this situation. Give me a recommendation.
Obstacle analysis
I am stuck on [OBSTACLE] in my project. Describe it: [WHAT IS HAPPENING]. Help me analyze this: Is this a real blocker or a perceived one? What have I already tried? What am I avoiding by staying stuck here? What are the unconventional options I have not considered? And what is the smallest action I can take right now to start moving, even if it does not fully solve the obstacle?
Priority reset
My project is in week [N] and I have had these developments: [DESCRIBE WHAT CHANGED]. My original priorities were [ORIGINAL PRIORITIES]. Given what I know now, what should my priorities be? What was important in week 1 that is less important now, what has become more urgent, and what I should stop doing entirely to protect the project's core outcome. Help me reorient without abandoning the plan.
Energy and motivation
I have the time to work on [PROJECT] but I am not doing it. The resistance feels like: [DESCRIBE: DREAD, BOREDOM, OVERWHELM, PERFECTIONISM, UNCLEAR NEXT STEP]. Be direct: What is actually going on here? Is this a plan problem (fix the plan), a skill problem (I do not know how to do this), an environment problem (something external is blocking me), or an avoidance problem (I am afraid of something)? Tell me which and what to do about it.
Scope creep response
My project has expanded from the original scope. Here is what was added: [LIST SCOPE ADDITIONS]. I need to decide what to do. Analyze: what drove this expansion (legitimate new information versus poor boundary-setting), what the impact is on timeline and quality, what I should cut elsewhere to accommodate the additions if I keep them, and whether I should push back on any additions and how to do that. Say yes to less but do more with what you say yes to.
Planning improves through honest review. These prompts build the retrospective discipline that makes each project more effective than the last.
Weekly retrospective
Week [N] of [PROJECT]. Planned: [WHAT I PLANNED]. Actual: [WHAT HAPPENED]. Run my retrospective: what the gap between plan and reality tells me (about my estimates, my priorities, or my execution), what specifically slowed me down, what I should do differently next week, and whether I need to update the milestone dates. Be precise, "do better" is not a retrospective finding.
Milestone review
I just hit milestone [N]: [DESCRIBE MILESTONE AND STATUS]. Review it against the original plan: did I hit it on time, at the quality I expected, at the resource cost I expected? What did this milestone teach me about the remaining milestones, are they still realistic? And what should I do with the lessons from this phase before moving to the next one?
Plan versus reality debrief
My project is complete (or I am abandoning it). Original plan: [DESCRIBE]. Actual result: [DESCRIBE]. Run a full debrief: the biggest gap between what I planned and what happened, the estimate that was most wrong and why, the decision I would make differently, the thing that went better than expected and what made it work, and the one thing I carry into my next plan. I learn more from the gap than from the success.
Estimation calibration
Here are the time estimates I made for my project and the actual time taken: [LIST TASKS WITH PLANNED AND ACTUAL TIMES]. Analyze my estimation patterns: do I systematically underestimate a category of work, where is the variance highest, and what rule of thumb I should apply to my estimates for similar work in future. Most people overestimate their speed at familiar work and underestimate unfamiliar work, tell me my specific bias.
System update from learning
Based on everything this project taught me, help me update my planning system: the template change I should make, the rule I should add to my pre-project checklist, the question I should always ask that I did not ask this time, and the default assumption I should stop making. Do not produce a long list, give me the one or two changes that would most improve how I plan the next project.
Claude tends to push back more directly on vague goals and weak reasoning, making it particularly useful for the goal-clarification and assumption-mapping work in stage one. It is also strong at structured reasoning, dependency mapping, pre-mortems, and decision analysis, where the rigor of the output matters as much as the content. For planning work that benefits from a thinking partner rather than just a task executor, Claude is a strong choice.
Yes. The backwards planning, dependency mapping, and resource plan prompts in stage two are designed for complex projects with multiple moving parts. The pre-mortem prompt is particularly valuable for high-stakes plans: imagining failure before it happens surfaces risks that forward-thinking misses. Claude can also hold a lot of context, which helps when you need to reason about a plan with many interdependent parts.
Specificity is everything. Instead of "help me plan to grow my business," try "help me plan to grow my consulting revenue from 120k to 180k in 12 months, with these current clients, this many available hours per week, and this constraint." The more real constraints and context you give, the more the output resembles a plan you can actually follow rather than a generic framework.
AI Prompts for Claude for Productivity
Claude's ability to reason through complex trade-offs and give nuanced, direct responses makes it well-suited for productivity work that goes beyond simple task completion.
See promptsAI Prompts for Claude for Business Planning
Build a rigorous, investor-ready business plan using Claude to research, challenge assumptions, structure strategy, and write every section..
See promptsAI Prompts for Claude for Writing
Claude is widely regarded as the best AI for writing quality.
See prompts