stop-doing-list

Stop Doing List — The Method for Building a Not-To-Do List

R — Original Text (Reading)

"A good company definitely has a long 'Stop doing list,' that is, 'things not to do.' The secret is not what you did but what you didn't do."

— Duan Yongping, Section 7

"When you discover something is wrong, you must stop immediately; no matter how great the cost, it is the smallest cost."

— Duan Yongping, Section 3

I — Methodological Skeleton (Interpretation)

The Stop Doing List is a reverse strategic framework that defines "what to do" through "what not to do." It is not passive avoidance, but active strategic exclusion.

Core logic:

Two types of not doing: universal taboos (what no one should do, such as deception) + mission binding (what conflicts with the company's vision, such as things that damage the brand)
Error-correction mechanism: when you find a mistake, stop immediately; at that point the cost is always the smallest—sunk cost is not a reason to continue
Cumulative effect: persist for 10-20 years, and the gap with opponents who do not have a not-to-do list will be enormous
Construction method: behind every not-doing item there is a story, a reason, and logic; it is not listed on a whim

Specific items in Duan Yongping's system (as a reference template): do not bargain, do not do OEM, do not borrow money, do not sell on credit, do not delay payment for goods, do not delay wages, do not do dishonest things, do not attack competitors, do not engage in price wars, do not pursue cost-effectiveness, do not make undifferentiated products, do not overtake on the curve, do not expand blindly, do not make quick money, do not exaggerate products.

A1 — Application in the Book (Past Application)

Case 1: BBK rejects OEM orders

Problem: Walmart placed an OEM order for 1 million VCDs, with rich profits
Use of the methodology: Duan Yongping asked, "Looking back 10 or 20 years from now, does doing OEM help us build our brand?" The answer was no—resources would be dispersed
Conclusion: reject the order and go all in on our own brand
Result: later OPPO/vivo became top five global phone brands, while most companies doing OEM at the same time disappeared

Case 2: OPPO exits the soy milk maker market

Problem: OPPO made soy milk makers and found the direction was wrong
Use of the methodology: when you discover a mistake → stop immediately, even at the cost of hundreds of millions in losses
Conclusion: exit the soy milk maker business and focus on phones
Result: after focusing, OPPO became a leading global phone brand

Case 3: unified pricing, no sales department

Problem: the industry convention is that large customers negotiate discounts and small customers pay full price
Use of the methodology: put "do not bargain" on the Stop Doing List
Conclusion: one price for all customers, eliminating the entire sales department and negotiation costs
Result: OPPO/vivo combined annual revenue of 40 billion USD, with no Sales department

A2 — Trigger Scenarios (Future Trigger) ★

In what situations would a user need this skill?

The user is formulating company strategy or a personal plan, has listed a bunch of things they want to do, but lacks screening criteria
The user's company is growing too fast, and they worry about losing control or making a fatal mistake
The user faces an opportunity that "looks great but has hidden concerns" and is hesitating about whether to pursue it
The user repeatedly makes the same type of mistake and wants a systematic error-correction mechanism
The user is thinking about "why some companies can endure while others are flashes in the pan"

Language signals (activate when these appear in the user's words)

"There are many opportunities, and I don't know how to choose"
"I want to try everything"
"It's growing too fast and I'm a bit worried"
"How do I avoid pitfalls/crashing"
"Why can some companies live a long time"
"We made a mistake but have already invested a lot"
"Should I do XX?"

Distinction from adjacent skills

Difference from right-things-first : right-things-first is the underlying philosophy of "judging whether the direction is right," while stop-doing-list is the execution tool that operationalizes it into a concrete list
Difference from business-model-evaluator : business-model-evaluator evaluates whether a business model is good or bad, while stop-doing-list lists "things that conflict with a good model and must never be done"

E — Executable Steps (Execution)

After the skill is activated, the agent should execute the following steps:

Help the user establish two types of Stop Doing List
- Completion standard: list at least 3 universal taboos + at least 3 mission-bound items
- Each item must be able to clearly explain the logic behind "why not do it" in one sentence
Validate each item with a "looking back after 10 years" test
- Completion standard: for each item, ask yourself, "Looking back after 10 years, is it right to have persisted in not doing it?"
- Stop condition: if any item does not hold up from a 10-year perspective, re-examine whether it should be included
Establish an error-correction mechanism
- Completion standard: clarify the specific steps for "how to stop once a mistake is discovered"—who has the authority to call a halt, when to call a halt, and the follow-up process after calling a halt
- Key principle: no matter how great the cost, it is still the smallest cost
Regularly review and supplement
- Completion standard: recommend that the user review the Stop Doing List every six months and distill new items from newly made mistakes

B — Boundary ★

Do not use this skill in the following situations

Daily task management (the Stop Doing List is at the strategic level; not every small thing needs to be listed)
When the user already has a clear action plan and only needs execution support (do not over-analyze)
Purely academic discussion or information lookup (no operationalization needed)

Failure modes the author warns about in the book

A list that is too long will tie your hands: "The list items should be as few as possible, otherwise they may tie your hands"
Turning Stop Doing into passive withdrawal from the world: not doing something is for the sake of doing something better, not for the sake of doing nothing
Surface-level execution without internalization: Duan Yongping said, "It is extremely difficult to carry corporate culture through to the nerve endings"

The author's blind spots / limitations of the era

Duan Yongping's Stop Doing List is built on the foundation that his enterprise already has profitability; for startups, appropriate adaptations may be needed (e.g., "don't borrow money" does not fully apply to startups that need financing)
Certain items (such as "don't attend trade shows") may be strategies specific to a particular industry and are not universally applicable

Adjacent methodologies that are easily confused

Difference from "risk management": Risk management assesses probability and impact, while the Stop Doing List draws red lines—no matter how low the probability, do not touch it
Difference from "subtraction": Subtraction is about optimizing efficiency, while the Stop Doing List is about preventing fatal mistakes—the purposes are different

Related skills (to be filled in Phase 3)

depends-on: right-things-first — stop-doing-list is an operational tool for the underlying philosophy of "doing the right things"; only after understanding "why direction must be judged first" can an effective not-to-do list be established
contrasts-with: (none)
composes-with: dare-to-follow — stop-doing-list says "don't make products without differentiation," while dare-to-follow says "only with differentiation can you enter later," the two sides complement each other

Audit information

Verification passed: V1 ✓ / V2 ✓ / V3 ✓
Test pass rate: {{%}} (see test-prompts.json for details)
Distillation time: 2026-04-16