Page edit frequency

Page edit frequency is the average number of edits made to existing pages over a defined period, expressed per page. If a 500-page knowledge base receives 1,000 edits in a month, the edit frequency is 2 edits per page per month. It can also be read as how often a given page is touched, which surfaces both heavily maintained pages and ones that have not changed in years.

The metric matters because content that is never edited quietly goes stale, and stale content erodes trust in the whole knowledge base. A healthy edit frequency means the team is keeping pages accurate as the product, policy, or world around them changes. A frequency near zero on important pages is an early warning that the knowledge base is drifting out of date even while it looks complete.

Edit frequency is the maintenance counterpart to page creation rate. Creation measures how fast you add content, edit frequency measures how well you keep it alive. The two together describe the real health of a knowledge base, in the same way that adding new customers and retaining existing ones together describe the health of a subscription base, a balance also seen in retention rate.

The base calculation divides total edits by the number of pages over a period. The definitions of an edit and the page scope decide whether the number means anything, so set them deliberately before you measure.

1. 1

   Total edits

   The count of edits saved in the period. Decide whether a meaningful edit is any save, or only a save that changes content beyond formatting, and apply that rule consistently.

2. 2

   Pages in scope

   The number of pages you are averaging across. Limiting scope to active or important pages gives a far more useful frequency than averaging across an archive that no one expects to change.

3. 3

   The period

   The window over which you count edits, such as a month or quarter. Keep it consistent so freshness trends remain comparable across periods.

4. 4

   Distinct editors

   The number of different people making edits. A frequency driven by many contributors is healthier than the same figure produced by one person reworking the same pages.

Worked example: a knowledge base of 400 active pages receives 600 content edits in a quarter from 15 distinct editors. The edit frequency is 600 divided by 400, which is 1.5 edits per page per quarter. But an average hides distribution. If 50 pages absorbed 500 of those edits and the other 350 pages were never touched, the average looks healthy while most of the base is going stale. Always look at the distribution behind the average, because the pages with zero edits are usually the ones the metric exists to catch.

An average edit frequency tells you how active maintenance is overall but not why, or where the gaps are. A metric tree decomposes the frequency into the forces that drive editing and traces each to the team or process that controls it. This turns a maintenance average into a map of where content stays fresh and where it rots.

The first level splits edit frequency into the demand for changes, the ease of making them, the clarity of ownership, and the share of content that is being neglected. Each branch decomposes further. Change demand reflects product releases, policy updates, and reported errors. Editing ease covers tooling friction and approval delay. Ownership clarity covers whether each page has a named maintainer and a review cadence. Neglect covers the count of stale pages and how long since they were last touched.

This structure lets you diagnose a change precisely. If overall frequency looks fine but trust is falling, the tree shows whether edits are concentrated on a few pages while important ones sit unowned and stale. Each diagnosis points to a different owner and a different fix.

There is no single benchmark for page edit frequency, because the right pace depends on how fast the underlying subject changes. A pricing or product page needs frequent edits, while a stable reference page may correctly go untouched for a year. The ranges below orient by content type, but the most useful benchmark is whether each page is edited as often as its subject actually changes.

Read these cadences against each page rather than the base as a whole. The most useful version of this metric is the share of important pages that have been reviewed or edited within their expected window, because that catches the specific pages drifting out of date. A single average across every content type will always look reassuring and tell you very little.

Improving edit frequency means getting the right pages edited at the right cadence, not lifting an average. The metric tree shows whether the constraint is unclear ownership, editing friction, missing demand signals, or a backlog of neglected pages, and that is where to focus first.

The metric tree approach starts by finding which branch is leaving content stale, then assigning a clear owner to fix it. Ownership and review cadence sit with whoever runs the knowledge base. Editing tooling sits with the platform team. Change signals sit with the teams shipping the products and policies that make pages go out of date.

KPI Tree fits this metric closely because its model puts named accountability on every node and pushes an alert to the owner when their branch moves. When a critical page crosses its staleness threshold or an unowned cluster of pages stops being edited, the accountable person sees it rather than the rot hiding inside a healthy average. The verified impact loop then checks whether assigning an owner or adding a staleness alert actually raised the edit frequency on the pages that mattered, so maintenance effort goes where it works.

1. 1

   Trusting the average over the distribution

   A healthy average can hide a base where a few pages absorb all the edits and the rest go stale. Always look at the distribution and the pages with zero edits.

2. 2

   Treating more edits as better

   Constant editing of the same page can signal churn or unclear ownership rather than good maintenance. Read frequency alongside what is being edited and why.

3. 3

   Averaging across all content types

   Fast-changing docs and stable reference pages have completely different healthy cadences. Mixing them produces a number that is true on average and useful for nothing.

4. 4

   Counting trivial saves as edits

   Counting every formatting tweak or whitespace save inflates the frequency without reflecting real maintenance. Define a meaningful edit and apply it consistently.

5. 5

   Tracking the frequency without decomposing it

   A headline edit frequency shows activity but not where content is rotting. Without breaking it into demand, ease, ownership, and neglect, stale pages stay invisible.