
Keep, update, merge or retire? Making decisions about an ageing blog
Turn an untidy content library into a clear action plan using business purpose, customer needs and evidence beyond page views.
Written by Rob KnottAn ageing blog often contains several different problems: useful advice that needs refreshing, overlapping articles, outdated offers and pages that were never connected to a business goal. Publishing more content does not resolve those decisions.
A content audit should give each page a supported action and an owner. This guide shows how to decide what to keep, update, merge or retire, with a fictional six-page example. It is designed for the person who needs a practical plan rather than another spreadsheet full of scores.
Start with the job each page is supposed to do
Before reviewing performance, write down the intended audience and purpose. A page might introduce a problem, help a customer compare options, support an existing client, explain a product or provide evidence for a sales conversation. These purposes need different success criteria.
Ask the people who use the content. Sales may repeatedly send a specialist guide that has little search traffic. Support may depend on an old setup article. A compliance or product owner may need a historical document to remain available. Low traffic does not make those pages worthless.
Keep the assessment grounded in the current business. A page promoting a service you no longer offer may attract visitors while creating confusion. Its popularity does not automatically justify keeping the promise unchanged.
Build an inventory you can make decisions from
Start with the URLs from your CMS, crawl and sitemaps, then reconcile them with search, analytics and backlink evidence where available. Include documents and old URLs if they form part of the content journey. Deduplicate carefully, retaining enough information to understand variants.
Record a sensible observation period and its limitations. A recently published page has had less time to accumulate evidence than an established guide. Seasonal content should be compared across appropriate periods. A tracking change can make apparent year-on-year differences unreliable.
Do not spend weeks assembling every possible metric before reading a page. A focused inventory with purpose, evidence and a decision is more useful than a comprehensive export that nobody can interpret.
| Field | Why you need it |
|---|---|
| URL, title and content owner | Identifies the page and who can approve changes |
| Audience, purpose and current offer | Connects the page to a real business need |
| Search and visit evidence, with dates | Shows discovery patterns within known limits |
| Enquiries or assisted use | Adds business context beyond page views |
| Links and references | Identifies dependencies and useful routes into the page |
| Accuracy and uniqueness | Shows whether the content deserves to remain |
| Decision, reason and destination | Makes the recommendation actionable |
| Owner, effort and verification | Turns the audit into a delivery plan |
Use four actions, with reasons
Keep a page when it remains accurate, useful and distinct. That decision can include routine maintenance; it does not mean the page should never be reviewed again. Update it when the purpose is still valuable but the information, examples, structure or route to action needs work.
Merge pages when they substantially serve the same need and a consolidated version would help the reader. Similar keywords alone do not prove duplication. Compare the actual questions answered, audience, search intent and business function. Two pages can legitimately address different stages of the same topic.
Retire content when it no longer serves a justified purpose and cannot sensibly be repaired or consolidated. Decide whether a relevant successor exists. A redirect should take the reader somewhere appropriate, not simply preserve a redirect count. Where there is no relevant replacement, a genuine removal response may be appropriate. Handle legal, contractual and customer-support retention needs before changing availability.
Worked example: six pages on a fictional B2B site
Imagine a training company that has accumulated several years of articles. It now sells structured team programmes rather than one-off public classes. The examples below are invented to illustrate decisions, not results from a client project.
The team reads each page, checks the available evidence and asks the commercial owner whether the offer is still valid. It then records a decision that another person can understand. Notice that neither high traffic nor low traffic determines the answer on its own.
| Page and evidence | Decision | Reason and next step |
|---|---|---|
| Current programme guide; used by sales and receiving relevant enquiries | Keep | Maintain factual details and preserve its useful enquiry path |
| Old assessment guide; useful purpose but outdated examples | Update | Refresh with a subject expert; retain the established URL if appropriate |
| Two guides answering the same buying question, each incomplete | Merge | Choose a suitable destination; combine useful material and review the old URL |
| Popular article advertising discontinued public classes | Update or retire | First decide whether it can honestly serve a current need; do not retain a misleading offer |
| Low-traffic customer setup guide linked from onboarding emails | Keep | It supports existing customers; coordinate with support before any move |
| Obsolete event announcement with no continuing purpose | Retire | Check retention and references; avoid an irrelevant homepage redirect |
Investigate overlap before declaring cannibalisation
If several URLs appear for related queries, inspect whether they actually compete to answer the same need. A category page, buying guide and support article may all be useful. The right response may be clearer positioning and internal linking rather than consolidation.
Look at query groups, page content and changes over time. Search data is incomplete, so a single query row should not decide the fate of an article. Read the pages together and write one sentence explaining what each should uniquely provide.
When you do merge, choose the destination based on relevance, usability, existing signals and maintainability. Preserve the strongest useful material, improve its organisation and update internal links. Record the old-to-new decision so the developer does not have to infer your intention.
Prioritise a manageable first batch
Start with confirmed inaccuracies, misleading offers and important customer journeys. Then choose improvements where there is a clear business purpose and enough evidence to act. Do not let the easiest cosmetic updates absorb the whole budget.
Estimate editorial and technical effort separately. Updating a paragraph may require a subject expert; merging pages may need redirects and changes to email links; retiring a download may require support approval. The owner of the spreadsheet is not necessarily authorised to make every decision.
Pick a small batch whose outcomes you can check. A ten-page pilot can reveal missing access, disagreements about ownership or flaws in your criteria before you apply the process to hundreds of URLs. Ten is a practical example, not a required audit size.
Write an actionable update brief
A recommendation such as “optimise this article” does not tell a writer what to do. Explain the reader's question, the weakness in the current answer, the evidence needed and the intended next action. Specify what should be retained as well as what should change.
For the fictional assessment guide, the brief might ask an expert to confirm the current process, replace obsolete screenshots, add a worked example and clarify who the service is suitable for. That is a more useful assignment than adding a target number of words.
Set an acceptance check: factual approval, complete links, correct destination, readable layout and a relevant next step. Avoid padding the article to satisfy a word-count target when a shorter explanation would serve the reader better.
Check the release and review the effect
Before changing URLs or availability, preserve a copy and the decision record. After release, test the destination, response behaviour, internal links and the customer journeys that use the content. Confirm that approved material actually reached the public page.
Review later performance using the same definitions and appropriate comparable periods. Look at relevant discovery, useful engagement and enquiries, while noting seasonal changes and other releases. A traffic increase is not automatically a commercial improvement, and a short-term decline does not establish that the decision was wrong.
Keep the evidence for decisions that did not work as expected. Content strategy improves when the team can revisit its reasoning, not when every completed ticket is presented as a success.
If you move or merge URLs, apply the redirect testing method to the affected routes.
Turn the audit into your next content plan
Use the gaps left by the audit to plan new work. You may discover that you already have plenty of introductory articles but little to help a buyer compare options or commission a project. That is a more useful starting point than copying every keyword a competitor ranks for.
A good handover includes a prioritised inventory, page briefs, dependencies and a realistic production sequence. If your team needs help making the decisions, you can commission the audit and strategy as a defined project, then keep delivery in-house or bring in additional support where needed.
Use the keyword research and content strategy framework to connect the audit with future demand.
Key takeaways
- Assess purpose and usefulness alongside search and conversion evidence.
- Keep low-traffic pages that still serve an important customer need.
- Merge only where a coherent replacement genuinely improves the answer.
- Turn every decision into an owned action with a verification step.

