When you update an old blog, start with posts that break a promise to the reader. A broken link, outdated fact, missing image, misleading title, or incomplete answer matters more than a dated font choice or an old screenshot that still explains the right thing.
The first goal of a content audit is not to make every page look new. It is to make the library trustworthy again.
Define the promise before judging the post
For each published post, write one sentence:
> A reader opens this post to understand or accomplish ___.
Then test the page against that promise.
- Does the title describe what the page delivers?
- Does the opening answer the question the title creates?
- Are factual claims still accurate?
- Do internal and external links reach the intended destinations?
- Do images load and still support the explanation?
- Can the reader take the promised next step?
This prevents a common audit mistake: rewriting a page because it feels old even though it still works, while a newer-looking page quietly sends readers to broken or irrelevant destinations.
Fix failures in risk order
Use four levels.
1. Harmful or misleading
Correct unsupported claims, obsolete instructions, expired offers presented as current, and guidance that could cause a bad decision. If you cannot verify a claim, narrow it, source it, or remove it.
2. Broken function
Repair dead links, missing media, forms that no longer work, and navigation that sends the reader to the wrong place. Links are a discovery and navigation resource for both people and search engines, so a broken path is more than a cosmetic defect.
3. Incomplete answer
Add the missing comparison, step, example, success signal, limitation, or next question. Google's people-first guidance asks whether someone will leave feeling they learned enough to help achieve their goal. Use that as a reader test, not as a reason to pad the article.
4. Presentation debt
Clean up formatting, old styling, awkward spacing, weak image metadata, and unnecessary clutter after the more important failures are handled.
Choose update, combine, redirect, or preserve
Not every old post needs the same treatment.
Update when the reader intent is still useful and the page can fulfill it with focused corrections.
Combine when two pages compete for the same reader and neither gives a complete answer on its own. Preserve the strongest material, choose one clear destination, and plan the redirect carefully before changing public URLs.
Redirect or retire only when the page no longer has an honest job and a better destination truly serves the same reader. Do not delete first and figure out the path later.
Preserve when the page still delivers its promise. An old publication date is not proof that the advice is wrong.
The same principle applies to any honest inventory: the value comes from what you can use, not from creating activity for its own sake.
Keep evidence for every decision
For each post, record:
- the reader promise;
- the defect you confirmed;
- the source or live check used to confirm it;
- the chosen action;
- the verification result after the change.
That record keeps a content audit from turning into a memory contest. It also makes it easier to stop after the useful work is done.
Start with the pages people depend on
If the library is large, prioritize pages that:
- receive meaningful traffic;
- support an offer or important next step;
- earn internal links from several pages;
- contain time-sensitive facts;
- answer a question central to the site's purpose.
Then work outward.
An old blog does not become trustworthy because every post has a fresh date. It becomes trustworthy when the pages keep the promises their titles make, the paths work, and the reader can still use the answer.
