Content decay: find and fix fading articles with Search Console

Use the term for a persistent decline from a page’s own earlier search performance, rather than every quiet day. An article can become less useful when its product references are discontinued, its instructions are outdated, or the questions readers need answered have changed. Competing answers and changes in demand or search presentation can also affect the observed results.

If a URL unexpectedly disappears, becomes inaccessible, or changes after a release, check the live page and its indexing status before treating the body copy as the problem. A content edit does not repair a broken destination or a page that cannot be served.

The Performance report shows clicks, impressions, click-through rate, and average position for your site in Google Search. You can group the data by dimensions such as query and page. Those observations help identify which searches are showing your existing content.

Start with an existing page that has lost visibility or clicks, then inspect the query groups behind the change. The report does not explain causation by itself or show every search in the market. Compare the page with its earlier performance and with related pages before deciding that the article needs a rewrite.

Choose an established article that matters to the business and has enough search history for a meaningful comparison. Locate the decline at page level before editing. A whole-site traffic change can hide very different patterns across individual articles.

Choose a comparison period that makes sense for the page. Two equal 28-day windows can be an editorial starting point, not a universal minimum. Look at the longer trend and, when relevant data is available, a comparable seasonal period. A winter boot guide attracting fewer searches in summer needs different interpretation from a year-round guide losing the same established queries.

• Record the exact page URL, date range, search type, and any country or device filters.
• Save the starting clicks, impressions, CTR, and relevant query groups.
• Note stock changes, paid campaigns, redesigns, and other events that may affect behavior.
• Decide what reader problem you expect an update to solve before editing.

Filter to the page, then inspect its queries. Read the live article alongside that list. Is the page answering the observed questions directly? Does it contain the relevant comparison or example? Is the query asking to shop while the page only offers an explanation?

You can also inspect the pages associated with an interesting query. If more than one URL appears, read them before deciding that something is broken. They may serve different needs. Use the keyword cannibalization check to compare their tasks and record whether the reader is reaching the right one.

These are diagnostic prompts, not automatic rules. A low CTR may reflect position, result features, intent, or how the page is presented. It does not prove that the title is bad or the article needs to become longer.

Turn the diagnosis into a short update brief. Suppose a fictional planter guide used to attract drainage questions but now references a discontinued product and gives only a vague explanation. The update could replace stale references with verified current options, explain their drainage features, and show which dimensions and accessories to check. That scope follows the observed problem instead of rewriting every section.

That is a stronger instruction than ‘optimize for drainage keywords.’ It identifies the missing answer, the evidence needed, and the scope. It also gives the reviewer something concrete to assess.

• Answer gap: add the missing explanation, example, or supported product detail.
• Preview gap: make the SEO title and description accurately describe the existing answer.
• Navigation gap: connect the relevant guide, collection, or product where the reader needs it.
• Page-assignment gap: clarify which existing page owns the question before adding another URL.

Use the content brief template to record the change. If you use prompt-based editing, request a focused revision, compare it with the saved article, and review the evidence again before applying it. Keep the title, H1, summary, and meta description aligned with the revised page.

Save a dated record of what changed and why. Include the affected sections and links. If you altered product facts, record the approved source. If you changed the search preview, save the old and new title and description so you can interpret later observations.

Check the live page after publishing. Verify the headings, visible copy, images, and destinations. A reporting chart cannot tell you that an inserted product link goes to the wrong item; that still needs editorial review. Check the route from the article to each recommendation with the internal linking workflow. Keep an existing URL when the page still serves the same purpose rather than creating a fresh duplicate solely to change its date.

Return to the same page and query scope after an appropriate observation period. Compare the measurements, but also inspect whether the queries now align with the intended answer. An increase in impressions alone can mean a wider set of appearances; it does not establish that the article generated more customers.

Keep the sample size in view. If a small page goes from one click to two, ‘clicks doubled’ is mathematically true but a poor basis for a confident strategy claim. Record both counts and the context. Do not promise a universal time to ranking or treat an uncontrolled before-and-after change as proof that one edit caused the result.

Search Console’s data documentation explains aggregation and reporting limitations. Use those limits when reading the table. The useful outcome is a better-supported next decision: keep the update, refine it, investigate another cause, or leave the page alone.