Content Audits
Know what you have before you decide what to build next. A content audit systematically inventories and evaluates every piece of content — revealing what exists, what works, what is outdated, and what is missing.
01 — TL;DR
Two sentences.
A content audit catalogues, categorises, and evaluates every piece of content within a defined scope — producing an inventory revealing what exists, a quality assessment identifying what is working or broken, and a gap analysis identifying what is missing relative to user needs. Without it, redesign projects are built on assumptions about what exists that a structured audit would reveal to be wrong.
The practice draws on information architecture and library science traditions. For design teams, it prevents building a navigation around the wrong content model and spending weeks writing content that duplicates articles nobody knew existed.
02 — When to Use
Apply this when…
When NOT to apply it
Skip it when the content scope is too small (fifteen articles do not need a formal audit), when a lighter sampling would provide sufficient signal, when a reliable current inventory already exists, or when the scope is undefined (produces an unwieldy dataset).
03 — How It Works
The mechanism
A content audit forces systematic examination of content usually encountered only incidentally. By cataloguing everything in scope, it makes the full landscape visible — revealing duplication, inconsistency, and absence that are invisible when content is encountered one item at a time.
A means, not an end
Every audit should conclude with three outputs: a prioritised list to remove/archive, a list to update, and a list to create — each ranked by impact. An audit filed without an action plan has done analysis but skipped strategy.
04 — Real Example
Intercom's documentation restructure and the 40% redundancy
When Intercom restructured their developer documentation, they began with a content audit. The inventory of 800+ items revealed approximately 40% were either duplicating content, referencing deprecated APIs, or covering unsupported use cases.
The team had planned to redesign navigation around the existing content. The audit showed that model was the wrong foundation. A consolidation sprint reduced the inventory from 800 to 480 before any navigation work began. The resulting structure was simpler because it was built on content audited for redundancy first.
05 — In the Wild
Test yourself & see real examples
No examples yet — be the first.
Spotted contradictory or outdated content that clearly was never audited — or content so well-curated it must have been systematically reviewed? Submit what you observed.
Seen Content Audit skipped in a real product? Help grow the evidence base.
06 — Common Mistakes
Where teams go wrong
07 — Variations & Related Principles
Connected ideas
Content audits sit at the intersection of content strategy, information architecture, and UX research.
The most important pairing is content audits with card sorting. A content audit without card sorting produces a well-inventoried set with no user-grounded structure. Card sorting without an audit may produce a structure for content that should not exist. Together: know what you have, decide what to keep, learn how users expect it organised.
08 — 10-Min Exercise
Run it right now
⏱ 10 minutes · Solo · No prep
1. Pick one specific content area — a help centre category, onboarding tooltips for one feature, or error messages on one form. Define the boundary tightly.
2. Rate each piece against three criteria only: Accurate (still true?), Clear (new user would understand?), Complete (fully addresses the user question?). Use Yes / Partially / No.
3. Count results. "No" on Accuracy is a priority update. "No" on Clarity or Completeness is a lower-priority improvement. Any user question with no content addressing it is a gap.
4. If more than 30% have a "No" on any criterion, you have found a systemic content quality problem warranting a full audit. You now have a baseline to compare against next quarter.