Sturgeon's Law

Law 19 / 20 Question Easy Answers

In any field most output is unremarkable, so judge a category by its best work and invest in selection, but treat "90%" as a figure of speech, not a measurement.

  • Adage or reasoning rule
  • 5 min read
  • Illustrative scenario
  • Reviewed by Ali Dezhami, November 8, 2026

The dilemma

Illustrative scenario

Ruth runs content for a regional garden-supply retailer with several hundred how-to articles built up over years. A consultant's audit lands on her desk with a blunt recommendation: "Sturgeon's Law applies; most of this is crud. Delete the bottom 90% by traffic and start again."

The traffic report seems to agree. A small set of articles on lawn care and tomatoes bring most of the visits, and hundreds of pages get almost none. But when Ruth looks closer, some of the quiet pages, such as a guide to pruning a particular hedge, have high conversion to product pages and are linked from customer service replies. Others are thin, out of date or duplicate each other.

The new editorial budget will cover either a mass deletion and a fresh start, or a slower review. Ruth has to decide what "crud" means for her business, who decides, and which measure stands in for quality when traffic clearly does not.

The short answer

Sturgeon's Law is the saying that "ninety percent of everything is crud." Science-fiction writer Theodore Sturgeon offered it in a 1958 magazine column, defending his genre against critics who judged all science fiction by its worst examples: most of every field is mediocre, so a category should be judged by its best work. In marketing it is a reminder that most ideas, ads and articles will be unremarkable, so quality depends on generating real alternatives and filtering them well.

How strong is the evidence?

Adage or reasoning rule

A memorable saying or reasoning rule, not an experimental finding. Treat it as a prompt to check, not as evidence.

This is an adage: a witty observation, not a measured base rate. Nobody counted 90% of anything; the number is rhetoric, chosen to make a point about fair comparison. Sturgeon himself called the line a "revelation" and used "Sturgeon's Law" for a different maxim, roughly that nothing is always absolutely so; later readers attached his name to the 90% line.

What survives scrutiny is a narrower, practical idea. In many creative fields, the share of truly strong work is small and hard to predict in advance, so teams that generate several genuinely different options and evaluate them against a clear standard tend to find better work than teams that polish one idea. That is a reason to invest in selection: briefs, quality criteria, testing and editing.

Two conditions matter. First, "crud" needs a definition tied to a goal; a low-traffic page can be excellent at its job. Second, the adage is about judging fairly, so it cuts both ways: your own output is not exempt.

Investigating an example

Illustrative scenario

A content audit that defined "crud" first

A home-fitness equipment brand plans to prune its blog. Instead of cutting by traffic alone, the editor writes down what each page is for: attract search visitors, help existing customers or support a sale. She then scores pages on three questions: is it accurate and current, does it do its job, and is it the best page on the site for that topic?

The result is less dramatic than the original plan. Some high-traffic posts are thin and are rewritten; some low-traffic setup guides are kept because customers and support staff rely on them; duplicates are merged into the strongest version. A smaller group of pages that fail every test is removed.

This is a teaching scenario, not a real company. The useful move was to define quality before filtering, so the 90% figure never became a deletion quota.

A teaching scenario. The people and numbers are invented; it is not a company case.

Choose your move

You are in the dilemma above. Which move would you make? Open each one to see its trade-offs.

A Delete the bottom 90% of articles by traffic, as the consultant recommends.

Fast and visibly decisive, but it removes useful low-traffic pages along with the weak ones, breaks links from support replies, and treats a figure of speech as a quota.

B Define quality by each page's job, score every page, then rewrite, merge or remove accordingly.

Slower and needs editorial time, but it keeps pages that quietly earn their place and removes the ones that do not. Results are harder to summarize in one slide.

C Keep everything and only write new articles from now on.

Avoids difficult calls, but outdated and duplicate pages keep confusing readers and competing with each other in search. The quality problem grows.

No move wins in every business. The law tells you what to check, not what to choose.

How to apply it

  1. Define quality for the goal. Write what a good ad, article or concept must achieve before you judge any of them, for example accuracy, distinct idea and a measurable job.
  2. Generate real alternatives. Ask for several genuinely different ideas, not cosmetic variations of one, so selection has something to choose between.
  3. Filter with a fixed standard. Score options against the written criteria, ideally with more than one reviewer, before testing with customers.
  4. Test the shortlist fairly. Compare finalists against a control on the outcome that matters, and record why the losers lost.
  5. Audit your own past work. Apply the same standard to what is already published; the adage is not only about other people's output.

What to measure

Share of ideas or pages that meet the written quality bar, outcome per piece (conversions, assisted sales, support deflection), and lift of tested winners over control. Guardrail: never delete or retire work on traffic alone; check links, conversions and support use first.

Worked example

Worked example: scoring 600 how-to articles by their job, not traffic alone (illustrative numbers)
GroupArticlesShare of visitsAction
High traffic, meets quality bar 40 55% Keep and update
High traffic, thin or outdated 30 20% Rewrite
Low traffic, does its job well 150 10% Keep
Duplicates of a stronger page 120 8% Merge and redirect
Low traffic, fails every test 260 7% Remove

Illustrative scenario, not data from a real company. The share removed here is well under 90%, and some pages that a traffic cut would have deleted are kept because they do a useful job.

Where it breaks

  • "90%" is not a statistic. Using it as a target or a quota turns a joke into a policy.
  • Quality depends on the goal. A page with little traffic can be the best answer for the few people who need it.
  • Volume is not the fix. Making more mediocre work raises costs and noise unless selection improves too.
  • The judge matters. Who sets the quality bar, and with what criteria, shapes what counts as "crud".
  • Ethics: dismissing a category, a community's work or a competitor's product by its weakest examples is the unfair comparison Sturgeon was objecting to.

What if you apply it incorrectly?

A marketing team decides that since "90% of everything is crud", it should produce ten times as many ads and let the algorithm find winners. Production costs climb, every ad is a small variation of the same idea, and the platform's tests split traffic so thinly that none of them reaches a clear result. The team ends up with more work, and still no idea which message matters.

Myth versus evidence

The social-video version

Ninety percent of your content is worthless, so delete most of it.

What the evidence supports

The 90% is rhetoric, not a measured rate; Sturgeon's point is to judge by the best work and to filter carefully against a clear standard.

Five-minute field assignment

Pick ten pieces your team published or launched this quarter. Write one sentence on what each was supposed to achieve, then mark whether it did, using the best data you have. How many meet your own bar, and what do the strongest ones have in common?

Checkpoint

Two questions: one about the idea, one about a decision. Answer both correctly to master this law. Questions are drawn from a larger bank and graded on our server; a retry brings different questions.

Evidence ledger

Every factual claim in this lesson, where it comes from, and how far it has been checked.

  • Theodore Sturgeon published the "ninety percent of everything is crud" line in a March 1958 column in Venture Science Fiction, defending science fiction against critics.

    Source cited, awaiting editor check Sturgeon, T. (1958, March). On Hand: A Book. Venture Science Fiction, 2(2), 66–69.

    Limit: Exact wording on pp. 66–69 and whether he voiced it earlier in talks are to be confirmed.

  • Sturgeon called the 90% line a "revelation" and used "Sturgeon's Law" for a different maxim, roughly that nothing is always absolutely so.

    Source cited, awaiting editor check Sturgeon, T. (1958, March). On Hand: A Book. Venture Science Fiction, 2(2), 66–69.

    Limit: Commonly reported; check against the original column before publishing.

  • The garden-supply retailer, the fitness brand and the 600-article table.

    Illustrative, not a factual claim

    Limit: Teaching scenarios; no real companies.

Further reading

  • Sturgeon, T. (1958, March). On Hand: A Book. Venture Science Fiction, 2(2), 66–69.

Which other law could explain this decision?

Editorial review: Ali Dezhami, November 8, 2026. Last substantive update: November 8, 2026. Our methodology