"What percentage of companies have outdated docs?" is one of those questions where a confident single number should make you suspicious. Nobody has surveyed every SaaS company's help center and graded it. So instead of inventing a precise figure, it is worth being clear about what can actually be measured, and what can only be reasoned.
What we can measure is documentation coverage at the level of individual product changes. We audited a large sample of merged pull requests across open-source SaaS repositories and checked whether each customer-facing change ever got a matching docs update. That gives a defensible, change-level picture, which is a much more honest way to answer the question than a made-up company-level percentage.
Direct answer
Effectively every SaaS company that ships regularly carries some outdated documentation, so the useful question is not "how many" but "how much." Measured at the change level in our audit of 29,917 merged pull requests across 514 repositories, 79.7% of customer-facing changes eventually got a docs update, 20.3% shipped with none, and 6.3% were never documented at all, even long after merge. In other words, about 1 in 5 customer-facing changes ships undocumented. Any company shipping at a normal pace accumulates this debt unless something actively closes the gap. See the full methodology in the State of Documentation Debt report.
Why "X% of companies" is the wrong number
A headline like "73% of companies have outdated docs" sounds authoritative and usually is not. Documentation freshness is not binary at the company level: a business does not have "good docs" or "bad docs," it has thousands of individual pages at different levels of accuracy. And no dataset grades every company's help center, so any such percentage is either a small self-selected survey or an outright guess.
The honest unit of measurement is the change. Every customer-facing thing you ship either gets documented or it does not. Aggregate that across a large sample and you get a stable, defensible picture of how common outdated docs really are.
What the change-level data shows
Across the sample, coverage was high but never complete, and the misses clustered exactly where you would expect: fast, small changes that shipped quietly.
- 79.7% of customer-facing changes eventually received a docs update.
- 20.3% shipped with no docs update at all, about 1 in 5 changes.
- 6.3% were still undocumented well after merge, effectively permanent debt.
- Only 25.6% of all merged PRs were customer-facing in the first place, so the debt hides inside a minority of changes that are easy to lose track of.
Why the gap is universal, not a competence issue
The undocumented slice is not random and it is not caused by bad writers. It is a structural gap in the workflow.
- The person who changed the product is rarely the person who owns the article.
- There is no automatic trigger connecting a merged PR to the docs it affects.
- Small, fast changes are the easiest to forget and often the most confusing to customers.
Because the cause is structural, the fix is structural too. High-coverage teams do not try harder to remember; they make the merge itself the trigger. For the mechanics, see How DocsKoala works.
How to move your own number
- Treat every merged customer-facing PR as the trigger to update docs, not a memory task.
- Draft the update from the diff so it is ready to review immediately.
- Keep a lightweight human approval so accuracy stays high as coverage rises.
That is the exact loop DocsKoala runs. If you want to see what it would cost to close the gap on your own repository, the pricing page lays out the trial and paid tiers.
Frequently asked questions
What percentage of SaaS companies have outdated documentation?
Effectively all of them carry some, because documentation freshness is graded per page, not per company. The defensible, measured figure is at the change level: in our audit, about 1 in 5 customer-facing changes shipped with no docs update, and 6.3% were never documented at all.
Is there a reliable statistic for company-level doc quality?
No credible one exists, because nobody audits every company's help center. Any "X% of companies" figure is usually a small survey or a guess. Change-level coverage, measured across many merged PRs, is the honest way to size how common outdated docs are.
Does high documentation coverage mean no outdated docs?
No. Even teams that document most changes miss a steady minority every release, so some outdated content is always present. The goal is to shrink and quickly correct the gap, not to assume it can ever hit zero by effort alone.
How can a team reduce its outdated docs?
Trigger updates from merged pull requests rather than memory, draft the change from the diff automatically, and keep a quick human approval step so accuracy holds as coverage improves.
Conclusion
- There is no honest single "percentage of companies" number; nearly all carry some debt.
- Measured at the change level, about 1 in 5 customer-facing changes ships undocumented.
- 6.3% are never documented at all, forming permanent debt.
- The cause is a missing trigger, not bad writers, so the fix is structural.
- See the full report, or close the gap on your repo with the DocsKoala trial.