Why "good enough" is the whole point of 80/20
Ernie Svenson · email · 2026-07-03 source ↗ Reader 🖥️
The real value of the 80/20 rule is not the input-output ratio but the permission it grants to stop at 'good enough' — the outcome-relative level a job actually demands — because the last increment of effort usually goes unnoticed, so the true discipline is the confidence to quit polishing rather than the diligence to do more.
practical · Expository/argumentative newsletter essay in a coaching, hortatory register, built around one central anecdote (over-customizing a new tool) and a metaphor (excess fat as opposed to nutrition), addressed to lawyers; closes with a product plug (the Inner Circle). — The load-bearing move is reframing the 80/20 rule from an input-output ratio into a psychological permission-to-stop, and diagnosing perfectionism as fear of criticism rather than high standards. 'Good enough' is explicitly outcome/audience-relative ('the level the job demands', 'what the moment called for'), and Svenson does supply a locator for it — cultivated judgment plus an outside read from a mentor or an AI — so this is not a naive 'just stop'. The unexamined premises are that the demanded level is reliably discernible and that the outsourced read (especially a machine one) is trustworthy.
Structure
| Part | Role |
|---|---|
| Intro — perfectionism overrides 80/20 | Names the failure mode: people who love '20% of the work gets 80% of the result' sit down and can't stop, so a one-hour job takes four; that is not the rule failing but perfectionism insidiously overriding it. |
| 'The extra work nobody asked for' — the last 20% is unnoticed | States the diminishing/unnoticed-returns claim, illustrates it with over-customizing a new tool ('None of it wins a case'), the 'excess fat in a meal' metaphor, and redefines 80/20 as deciding the level of good-enough the job demands. |
| '“Good enough” isn't lazy. It's a target.' | Pre-empts the lawyer's objection that good-enough means cutting corners: it is instead thoughtful, strategic, focused work. |
| 'A lot of “perfectionism” is really fear' | Reframes over-polishing: sometimes love of precision, but often fear of being criticized — lawyers are trained to spot attackable flaws and brace for critics; that is a confidence problem, not a standards problem. |
| 'Why confidence is part of the equation' | The constructive core: knowing what to leave out takes confidence; novices can't tell load-bearing parts from decoration so keep it all; experience lets you discern what matters and cut — and you can borrow confidence via a mentor or an AI sounding board that flags what doesn't earn its keep. |
| 'Bottom line' | Fuses the 80/20 rule ('where the results come from') with 'the perfect is the enemy of the good' ('when to quit'); the work is not doing more but having the confidence to stop at good enough. |
Key terms
- good enough
A deliberate target: the quality level the essential part of a job actually demands, fixed by outcome and audience ('what the moment called for'), past which further effort is unnoticed waste — explicitly NOT laziness or cutting corners. - the 80/20 rule (Pareto principle)
Reframed from an input-output ratio ('do the 20% of the work that gets 80% of the result') into a permission and decision to STOP at good enough — 'one tells you where the results come from, the other tells you when to quit.' - the last 20%
The marginal effort past good-enough that 'usually produces something no one will notice' — over-polish, decoration, 'excess fat' that chews up valuable time. - perfectionism
Not high standards but usually fear of criticism in disguise — a confidence problem that insidiously overrides 80/20 and wastes enormous time. - confidence
The cultivated capacity to tell load-bearing parts from decoration and cut with conviction — buildable by experience or borrowable from a mentor or an AI; the real bottleneck to stopping at good enough.
Propositions
- explicitPerfectionism, not a flaw in the 80/20 rule, is what stops people who intend to 'do the 20% that gets 80%': they keep polishing and adding until a job that should take an hour takes four.conf: high
- explicitcontestedThe last 20% of effort usually produces something no one will notice.conf: highcontradictsYou can't unit test for taste · p12 — Same referent, opposite sign, on the past-adequacy increment. Svenson (p2/p9): the effort past 'what the job demands' usually produces something no one will notice, so it is waste — a consequentialist, audience-relative stopping rule ('None of it wins a case'; 'what the moment called for'). Tryggvason p12: 'correctness is not quality' — a result that clears the checkable/adequacy bar can still be tasteless, so passing that bar gives false confidence about quality, which lives in exactly the increment Svenson discards. Note that 'meeting the job's demand' is hitting a spec = the correctness dimension p12 says is not quality. BOUNDARY: Svenson holds for throughput/commodity work with a discoverable demand bar (his tool-config example, routine deliverables); the taste cluster holds where the differentiator is the un-specifiable increment (craft, design, prose, brand, high-stakes work). Reconciliation lever: Svenson's own load-bearing-vs-decoration discernment (p6) is the bridge — the residual disagreement is only his empirical prior that taste-critical cases are the minority.
- explicitThe whole point of 80/20 is deciding, for the most essential part of a job, the level of 'good enough' that the job demands.conf: high
- explicit'Good enough' is not laziness or cutting corners but a deliberate, thoughtful, strategic and focused target.conf: high
- explicitMuch of what looks like perfectionism is really fear of being criticized — a confidence problem, not a standards problem — and it wastes enormous amounts of time.conf: high
- explicitKnowing what to leave out is a skill built by experience: novices cannot distinguish load-bearing parts from decoration so they keep everything, while experience lets you discern what actually matters and cut with confidence.conf: highsupportsYou can't unit test for taste · p13 — Same faculty, described from two directions. Svenson p6: the core skill is 'knowing what to leave out' — telling load-bearing parts from decoration and cutting with confidence. Tryggvason p13: taste is exercised largely as restraint and proportion — what to filter out — so 'quality is as much about what you leave out as what you include.' Both locate the discernment in knowing-what-to-omit. Pairing this with the p12 contradiction gives the honest richness: same faculty, opposite default stopping rule (Svenson points it at stopping to save effort; the taste cluster points it at iterating toward the un-testable right).
- explicitcontestedYou can shortcut the wait for confidence by getting an outside read — a mentor, or an AI you hand a draft to — to tell you where 'good enough' is and where you are adding things that don't earn their keep.conf: highcontradictsYou can't unit test for taste · p16 — Svenson p7 recommends outsourcing the 'where is good enough / what doesn't earn its keep' judgment to an AI as a useful outside read. Tryggvason p16 holds that trusting an LLM's latent 'taste' is itself an unverifiable taste judgment — an act of faith that may launder the model's training-distribution bias (the anglophone-Wikipedia bias, the doubted Reykjavik=10). They clash on the same axis: is delegating the taste/adequacy judgment to a model sound? Svenson's 'sounding board, not a replacement' hedge narrows the clash but does not remove it — he never addresses that the AI's read on what to cut is unauditable or that leaning on it to decide 'when it's done' is the corpus's cognitive-surrender risk.
- explicitThe 80/20 rule and 'the perfect is the enemy of the good' are the same idea: one tells you where results come from, the other when to quit; the real work is having the confidence to stop at good enough.conf: highsupports#35 Lessons learned from a 2x Unicorn Founder (Uri Levine) · p8 — Two framings of one maxim. Levine p8: 'the biggest enemy of good enough is perfect — you win markets by being good enough, not perfect' (market-strategy version). Svenson p8: the 80/20 rule and 'the perfect is the enemy of the good' are the same idea, and the work is the confidence to stop at good enough (individual-craft / psychological version). Both assert good-enough-beats-perfect; note both extracts also carry the decision-making concept.
- implicitSvenson's stopping rule is outcome/audience-relative: 'good enough' is fixed by external consequence — whether the work 'affects a client matter', 'wins a case', is 'what the moment called for' — so effort beyond the point of consequence is by definition waste, and quality is treated as instrumental to outcome rather than intrinsic.conf: high
- implicitThe scheme presumes the marginal value of extra effort is discernible enough — by cultivated judgment, even without a measurable test — that a person can reliably locate 'good enough' and recognize the remainder as decoration.conf: medium
- implicitThe judgment of 'what doesn't earn its keep' can be soundly outsourced to an outside reader — a mentor or an AI — with that reader's taste taken as trustworthy; the risk that a borrowed (especially machine) read is itself unverifiable, bias-laden, or corrosive of one's own judgment is not considered.conf: medium
Arguments
You should stop at 'good enough' rather than chase the last 20%, because the marginal effort is usually wasted. strength: high
Explicit premises:
- The last 20% of effort usually produces something no one will notice.
- That extra work does not affect a client matter or win a case; it just chews up valuable time.
- The point of 80/20 is to decide the level of good-enough the job demands.
Implicit premises:
- The value of effort is measured by its effect on the outcome (client matter, case, the moment's demand) — a consequentialist, audience-relative standard of quality.
- Returns to effort actually diminish for the task at hand — the value curve saturates before the extra 20% is spent.
- You can tell, in advance and reliably, which parts are load-bearing and which are decoration.
Inferences:
- The argument holds wherever the demand bar is discoverable and value saturates (throughput/commodity work, the tool-config example, routine deliverables).
- It fails wherever the differentiating value is the un-specifiable last increment (craft, design, prose, brand, high-consequence legal work), because there the 'returns diminish' premise is false and the audience's real standard may be near-perfection.
Over-polishing is usually a confidence problem, so the cure is to build (or borrow) the confidence to stop, not to raise your standards.
Explicit premises:
- People often over-polish out of fear of being criticized, not because the work needs more.
- Knowing what to leave out takes confidence.
- Confidence can be built by experience or borrowed via a mentor or an AI sounding board.
Implicit premises:
- The outside reader (mentor or AI) has trustworthy taste about where 'good enough' really is.
- Borrowing someone else's judgment about when you are done builds, rather than substitutes for, your own discernment.
Inferences:
- If the borrowed reader's taste is itself unverifiable or biased — as it is for an LLM — the cure imports a new, hidden risk; and habitually letting an outside read decide 'when it's done' is exactly the dependency the corpus calls cognitive surrender.
Narratives
- Getting a shiny new tool and spending hours customizing every setting and color-coding folders to build the 'perfect' system — before using it even once — which affects no client matter and wins no case, just chewing up valuable time; framed as 'excess fat in a meal that's supposed to deliver just nutrition.'
→ The central illustration of the last-20%-as-unnoticed-waste claim, and the clearest case for the outcome-relative standard ('None of it wins a case') (supports p2, p9). - The experienced lawyer learning 'which details a judge will actually care about', where the novice keeps everything 'just in case' because they cannot yet tell load-bearing parts from decoration.
→ Illustrates discernment/confidence as a product of experience — the basis for cutting with conviction (supports p6). - Handing a draft to an AI at 11 p.m. and asking it to critique your thinking, poke holes in your argument, or tell you where you're adding stuff that doesn't earn its keep — 'a sounding board available... when no one else is', though it 'won't replace a good mentor.'
→ Illustrates borrowing an outside read to locate 'good enough' (supports p7); also the hook for the taste/cognitive-surrender tension — the piece recommends outsourcing precisely the taste judgment the corpus finds AI cannot be trusted with (g1).
Questions addressed
- ✅ Why do people who love the 80/20 rule fail to actually apply it?
Because perfectionism — often fear of criticism in disguise — insidiously overrides it, so they keep polishing past the point of value. - ✅ Isn't 'good enough' just an excuse for laziness or cutting corners?
No — it is a deliberate, thoughtful, strategic target; skipping the unnoticed last 20% is discipline, not sloth. - ✅ How do you know where 'good enough' is, and what to leave out?
Through experience-built confidence to distinguish load-bearing parts from decoration, accelerated by an outside read from a mentor or an AI. - ⭕ When is stopping at good enough the wrong call — when is the last, un-specifiable 20% actually the whole point?
Not addressed; Svenson asserts the last 20% is generally unnoticed and never marks the taste/craft/high-stakes domains where that increment is precisely the value. - ⭕ What are the risks of outsourcing the 'is it done' judgment to an AI?
Not addressed; he recommends it as a useful sounding board without confronting AI taste's unverifiability, its training-distribution bias, or the dependency it can create.
Gaps
- incompleteSvenson prescribes outsourcing the 'is it done / what doesn't earn its keep' judgment to a mentor or an AI ('the fastest way to build confidence is to borrow someone else's'), but never addresses that an AI's 'taste' is unverifiable and re-encodes its training-distribution bias (the taste cluster's central empirical finding), nor the cognitive-surrender/dependency risk of borrowed confidence — the exact move the corpus flags as the point at which your own judgment stops forming and the outside read becomes your read. The essay recommends the very outsourcing its neighbours warn against.
Read next: Read extract-you-cant-unit-test-for-taste (AI 'taste' is untrustworthy and launders anglophone-Wikipedia bias; the doubted Reykjavik=10 score) and the corpus's cognitive-surrender material on 'borrowed confidence'; ask when an outside read cultivates your own discernment versus when it substitutes for it and creates dependence on a non-stationary, biased arbiter. - incompleteThe claim that 'the last 20% usually produces something no one will notice' is stated as a near-universal and paired with an outcome-only standard ('wins a case'), but the piece never marks the domains where the last, un-specifiable increment IS the value — craft, design, prose, brand, and high-consequence work (a safety-critical filing, an appellate argument) where refusing to stop is warranted rigor, not fear. Without a boundary, the advice over-generalizes from routine deliverables to work where 'good enough' forfeits the whole point.
Read next: Read the taste cluster — extract-you-cant-unit-test-for-taste (correctness is not quality) and extract-how-to-write-with-style (Vonnegut's 'have the guts to cut' aims at what illuminates the subject, a taste target, not at good-enough) — plus Ousterhout on design quality; test whether 'good enough' is ever the wrong stopping rule and specify the tasks where it is. - incomplete'Good enough for whom' is left under-specified. The standard is audience-relative ('the level the job demands', 'which details a judge will actually care about'), yet Svenson treats the demanding audience as singular and self-evident. He does not address conflicting audiences, or the case where the relevant reader's genuine standard is near-perfection (a regulator, an appellate court, a brand, an adversary looking for the attackable flaw) — which would collapse his effort-versus-value gap and vindicate the 'perfectionist' he diagnoses as merely fearful.
Read next: Read on satisficing versus maximizing (Herbert Simon) and stakeholder / quality-attribute analysis: who sets the adequacy bar, and what the stopping rule should be when the audience is adversarial or the cost of a missed detail is asymmetric.
Related documents
- #35 Lessons learned from a 2x Unicorn Founder (Uri Levine)
- You can't unit test for taste
- Podcast: Will Agentic AI Bring Fantasia's Sorcerer's Apprentice to Life?: A Conversation with Tracy Bannon
- The best response to AI slop, infinite advice, and online noise is from Robin Williams
- AI Is Undermining Leaders’ Judgment. Here’s What to Do About It.
- Brain Food: Building Trust in Yourself Every Day
- Det skal være vanskelig
- The Hidden Realities of AI Adoption
- How a Museum Marketing Team Used AI to Bring People Closer to Art
- Brain Food: Intelligence vs. Wisdom