Turn a messy task list into a ranked plan
Feed in a messy task list and constraints and get a ranked plan with an explicit cut for what waits until next week and the reasoning behind it.
Last updated
Fill in the blanks
Assembled in your browser — nothing you type is stored or sent anywhere.
Your prompt
Model-agnostic — works in Claude, ChatGPT, Gemini or any AI chat.
Why this prompt works
- →Requiring a constraint-tied reason for every ranking, rather than just a reordered list, makes the plan defensible and lets the reader spot-check the logic instead of taking the order on faith.
- →Forcing an explicit line between this week and "Not this week" prevents the most common failure of prioritization prompts: a ranked list that never actually says what gets dropped.
- →The instruction to flag genuine ties instead of forcing an arbitrary order is the escape hatch — without it the model invents a tie-breaking reason that sounds decisive but isn't actually supported by the input.
Part by part
- a one-line reason tied to the constraints
- Makes the ranking arguable. A list without reasons cannot be corrected, because there is nothing to disagree with.
- rather than a restatement of the task itself
- Names the default filler. "Important because it is important" is what a reason field produces when the instruction is not explicit.
- Draw a clear line under the tasks that fit this week
- The cut is the actual output. A ranked list with no line still leaves you deciding where to stop, which was the original problem.
- Not this week
- A named section for what did not make it, with the reason. Deferred work that is written down gets revisited; work that silently drops off does not.
- say so explicitly instead of arbitrarily ranking one above the other
- Prevents false precision. Two genuinely equal tasks presented in an order implies a judgement that was never made.
Common mistakes
Prioritise these tasks.Without constraints there is no basis for a ranking, so the model applies generic urgency heuristics. The result looks decisive and encodes none of what actually matters this week.
Which of these should I do first?Answers for one task and leaves the rest unordered. The useful output is the whole sequence plus the cut, because the second and third decisions are the ones that shift when something slips.
Sort these by importance.Importance alone ignores dependency and deadline, so a task blocking three others ranks below a more significant one that nothing waits on. Sequencing and importance are different questions.
Questions
What constraints are worth stating?
Hours actually available, fixed deadlines, and what depends on someone else. Those three drive nearly every ranking decision, and without them any order is guesswork wearing confidence.
Why does the cut line matter more than the ranking?
Because the ranking alone still leaves the hard decision unmade. Naming what will not happen this week is the part that has to be communicated, and the part people avoid writing down.
Should I follow the order exactly?
Treat it as a proposal you edit. The model cannot see which task you have been avoiding, which one has a political deadline, or which will take three times its estimate — but it can produce a defensible starting order in seconds.
What if everything is urgent?
Then the constraints are missing something, usually the real deadlines. If a genuine tie remains after that, the prompt says to name it — which puts the decision back where it belongs rather than disguising it as analysis.