Deciding What Not to Work On When Everything Looks Important
A three-step way to prune a crowded list, using Peter Drucker's planned abandonment, Edward de Bono's PMI tool, and Jason Fried's default of saying no.
The list has forty items on it and every one of them has a reasonable person behind it. You have a free afternoon in three weeks. Ranking the list has been tried, twice, and it turns out that "high priority" is a category with thirty members.
Why does ranking the list never fix it?
Because ranking assumes you will do everything eventually, and the order only matters if that is true. It is not. The real constraint is attention, and attention does not expand when the list is neatly sorted.
Ask an assistant to prioritize forty items and you get a competent matrix: impact versus effort, urgent versus important. It will rank all forty. It will not tell you that fifteen of them should not exist, because that conclusion requires knowing your strategy and your appetite for disappointing people.
The following three steps are about stopping, not sorting.
Step one: which things would you not start today?
Peter Drucker's work on management, including The Effective Executive (1967), argues that effectiveness depends on concentration and that organizations accumulate activities that nobody reviews. His remedy is a regular, deliberate abandonment of what no longer produces results. The question he suggests asking of any existing effort is whether, knowing what you know now, you would start it today.
Applied: Take the list and mark everything already in progress. For each, ask the question honestly. For items that fail, write down the cost of stopping and who needs to be told. Do this before adding anything new.
"Here is my current list of commitments and projects: [paste]. For each, ask me whether I would start it today knowing what I now know, and what the real cost of stopping would be. Then group the list into three buckets: clearly keep, clearly stop, and unsure. For the unsure bucket, suggest the one fact that would settle each."
Why it works: asking whether you would start something today separates its value from the sunk effort already spent.
Step two: what does a fair look at the unsure ones show?
Edward de Bono's CoRT thinking program introduced PMI: list the Plus points, the Minus points, and the Interesting points of an idea before deciding. The third column matters. It collects consequences, questions, and possibilities that are neither clearly good nor bad.
Applied: For each item in the unsure bucket, spend three minutes on a PMI. Interesting points often reveal that an item matters for a reason you had not named, such as it opens a relationship, or that it matters for no reason at all. Decide after the PMI, not before.
"For each of these items: [list], run a PMI. List the Plus points, the Minus points, and the Interesting points, meaning consequences or questions that are neither clearly good nor bad. Then tell me which items look weaker after the exercise and which look stronger, and say what evidence would change your view."
Why it works: separating the three columns shows what you have been assuming without checking.
Step three: how do you keep the list short once it is short?
Jason Fried and David Heinemeier Hansson's Rework (2010) argues for saying no by default. Their reasoning is that every feature, project, or commitment adds ongoing weight, and that waiting often shows whether an idea really matters. If something is important, it will return.
Applied: Create a "not now" list and write down each declined request with the date. Review it monthly. Most will stay there. The few that keep returning have earned a conversation. This also makes saying no easier to defend, since the idea has a place to wait.
"Help me set up a "not now" list for requests I decline, with a monthly review routine. Then draft a short, respectful reply I can send when I decline a request, one that says what I am focused on, that the idea has been recorded, and when I will revisit it."
Why it works: a place for declined ideas lowers the cost of saying no for you and for the person asking.
Which step fits where you are?
| Approach | From | When it fits | Watch out for |
|---|---|---|---|
| Planned abandonment | Peter Drucker, The Effective Executive | The list is full of things already underway that nobody reviews | Stopping things that look weak but are quietly load-bearing |
| PMI | Edward de Bono, CoRT thinking program | You cannot tell whether an item deserves its place | Using the exercise to rationalize a decision already made |
| Default no | Jason Fried and David Heinemeier Hansson, Rework | New requests keep refilling the list after you cut it | Declining things that carry a real obligation to someone |
What won't pruning solve?
It cannot tell you your strategy. If you do not know what you are trying to achieve, you cannot judge what to drop, and every cut will feel arbitrary. It also cannot remove commitments that are genuinely not yours to cancel, such as a customer contract or a duty to a team. Those need a conversation, not a list.
Peter Drucker is in the Operator & Systems category, Edward de Bono in Educator, and Jason Fried in Bootstrapper. For the originals, read The Effective Executive, de Bono's CoRT materials, and Rework. For a structured look at a decision once it is narrowed down, Decision Brief is a $79 tool built for high-stakes choices.
All the copy-paste prompts from this guide, and the rest of the Situations series, live at /prompts too, free to copy without reading the guide first.
Frequently asked questions
How is this different from ordinary prioritization?
Prioritizing ranks what you will do. Pruning decides what you will stop or never start. Drucker's point is that most lists only grow because nobody schedules the stopping. If you rank twenty items and still try to do them all, you have organized the overload without reducing it. The cut is the work. Set a deadline for the cut, such as this week, since the pile otherwise grows back faster than you can sort it.
What is planned abandonment?
It is Peter Drucker's term for deliberately ending activities that no longer earn their place, rather than letting them continue by default. His test is to ask, of any existing activity, whether you would start it today knowing what you now know. If the answer is no, the question is how to stop, not whether. Be honest about commitments you cannot drop, and flag them in a separate list so they stop competing with real choices.
What is a PMI, and why use it instead of a pros and cons list?
PMI stands for Plus, Minus, Interesting and comes from Edward de Bono's CoRT thinking materials. The Interesting column captures consequences and possibilities that are neither clearly good nor bad, which a plain pros and cons list tends to miss. It slows a gut reaction long enough to see more of the option. Spend a few minutes on each column, and note which Interesting points deserve a real follow-up question before you decide.
Doesn't saying no by default risk missing opportunities?
It does, and Fried and Heinemeier Hansson accept that cost in Rework. Their reasoning is that every yes carries ongoing weight, and good ideas that matter will come back. A default of no turns each request into a choice rather than an obligation. Keep a list of declined ideas so you can revisit them deliberately. Treat the default as a starting position, not a vow, and keep a short list of the exceptions that you have explicitly chosen.
What if my boss or customers keep adding to the list?
Make the tradeoff visible. Show the list, say what would have to stop for the new item to start, and ask which they would choose. Drucker's approach is to treat capacity as finite, so every addition names a displacement. It turns a request into a decision they share. If they decline to name a tradeoff, you have learned that the request is a preference, not a priority, and you can say so politely.
Can AI decide what I should drop?
It can help structure the inventory, run PMI analyses, and challenge your reasons. It cannot know the politics, commitments, or strategy behind your list. Use it to expose assumptions, then make the call yourself. Give it your goals and constraints along with the list, and ask it to challenge each item you are tempted to keep. Treat its challenges as prompts for your own judgment, since only you know which commitments carry obligations the list does not show.
Written by Gareth Hoyle. Last updated 8 October 2026. Part of the authority.md guides library.
More guides.
Apologizing to a Customer When You Got It Wrong
How to write an apology a customer believes, using the Trust Triangle from Frances Frei and Anne Morriss, the Trust Equation from Maister, Green and Galford, and Robert Cialdini's commitment principle used ethically.
Asking for a Raise Without Turning It Into a Standoff
How to run a pay conversation as a problem you and your manager solve together, using three documented negotiation approaches from Fisher and Ury, Chris Voss, and Deepak Malhotra.
Running a Post-Mortem Without Blame That Still Changes Something
How to run a review after a failure that finds real causes without scapegoating anyone or letting accountability disappear, using Amy Edmondson, W. Edwards Deming, and Gene Kranz.