SEO
Refreshing old content before writing new
Decide which old pages deserve a refresh, scope useful changes, protect existing value and measure search results alongside editorial effort.
OppRise Editorial · Reviewed · 8 min read
The short answer
Refresh existing content before writing new when it already serves the right reader task but contains fixable gaps, inaccuracies or outdated instructions. Diagnose the problem, scope the changes and measure results. Create a new page when the audience or task is materially different, not simply to publish more.
What to take away
- Prioritise demonstrated reader problems, not old dates alone.
- Diagnose search changes before rewriting copy.
- Preserve useful content and URLs where the purpose remains unchanged.
- Measure search outcomes and editorial efficiency separately.
- Treat period comparisons as evidence of change, not proof of causation.
1. Start with the reader’s unfinished job
An old publication date is not a content defect. A page deserves attention when it no longer helps its intended reader make a decision or complete a task. Start by writing one sentence: “After reading this page, someone should be able to…” Then test the existing content against that promise.
Hypothetical example: a removalist’s packing guide still explains wrapping glassware correctly, but omits apartment lift bookings and access restrictions. Adding a practical access checklist may solve a real problem. Rewriting every paragraph to sound newer probably will not.
Google’s people-first content guidance supports improving usefulness and reliability rather than changing dates to create an appearance of freshness. Keep accurate explanations; replace unsupported claims and missing decision support.
- Name the reader, their decision and the information needed.
- Identify specific inaccuracies, omissions or difficult steps.
- Write new content if the proposed task genuinely differs from the existing page’s purpose.
2. Build an inventory you can actually act on
Avoid beginning with a full-site rewrite programme. Build a manageable shortlist from important services, commercially relevant guides, pages with declining search clicks and questions repeatedly raised by customers. Record the URL, reader task, business relevance, evidence of a problem, owner and proposed action.
Hypothetical example: a bookkeeping firm has an evergreen record-keeping guide and several old deadline announcements. The guide may merit an expert review; expired announcements may need an archive label rather than expansion. Low traffic alone does not establish that either is useless.
Assign one action: keep, refresh, consolidate, archive or investigate. Add a confidence field so a suspected tracking problem does not become an unnecessary editorial assignment. Check technical access and measurement before blaming the copy.
- Separate essential maintenance from growth opportunities.
- Record the evidence date and where it came from.
- Park pages without a clear reader benefit or accountable owner.
3. Diagnose the decline before editing
In Search Console, inspect the individual page and relevant query groups. Compare clicks, impressions, click-through rate and average position using consistent search type, country and device filters. Google’s performance report documentation explains that aggregation affects reporting and average position is an average, not a fixed ranking.
Hypothetical example: a garden centre’s pruning guide loses clicks after its seasonal peak. If impressions also fall while positions remain broadly similar, investigate seasonal demand before commissioning a rewrite. If impressions hold but click-through rate falls, inspect the query mix, title and search results rather than assuming the article lacks depth.
Check comparable seasonal windows where available. Site migrations, indexing issues, changed search results and measurement changes can complicate the picture. A broad average can conceal strong performance for one intent and weak performance for another.
- Save the filters, dates and baseline export.
- Separate branded queries from non-branded discovery where useful.
- Document competing explanations instead of declaring a cause.
4. Prioritise by useful improvement per unit of effort
Prioritise pages with a clear audience need, relevant demand, a specific fix and a realistic review path. Estimate the whole job: research, expert input, editing, design, publishing and quality assurance. A short article requiring legal review may cost more effort than a long practical guide.
Hypothetical example: a commercial cleaning company can clarify an existing office-cleaning scope page using an approved service checklist. A proposed infection-control article needs specialist verification. The first may be a better near-term refresh, while the second deserves a separate, properly resourced brief.
Do not turn an impact score into a forecast of extra leads. Use qualitative ratings with reasons and confidence. Give incorrect or potentially harmful information priority even when its traffic is low.
- High priority: material error, important task, available evidence.
- Investigate first: unexplained decline or uncertain search intent.
- Defer: cosmetic rewriting with no defined reader benefit.
5. Write a change brief, not a request to “improve SEO”
A refresh brief should specify what stays, what changes and what evidence is needed. Preserve useful specificity, clear definitions and sections that already answer relevant queries. Give the editor a bounded assignment instead of encouraging a complete rewrite.
Hypothetical example: a software implementation guide explains setup well but fails to identify who supplies the data. The brief could add a responsibility table, replace obsolete screenshots and clarify the handover checklist. Adding a long history of the software category would dilute the task.
Include acceptance criteria: every screenshot matches the current interface, responsibilities have an owner, and the reader can identify the next step. Set an effort limit; if research reveals a different audience or task, reconsider whether a new page is warranted.
- Keep: accurate sections and distinctive expertise.
- Change: named gaps, stale facts and confusing instructions.
- Verify: claims, examples, dates and subject-matter approval.
6. Add decision support instead of word count
Useful additions often include eligibility rules, exclusions, preparation steps, worked examples and explanations of tradeoffs. Put the main answer near the beginning, then supply enough detail to apply it. Longer is not automatically more useful.
Hypothetical example: a furniture retailer’s dining-table guide could explain how to measure the room, allow for chairs and check delivery access. It should verify actual product dimensions rather than inventing a universal clearance rule. A diagram and measuring checklist may outperform several extra paragraphs as reader support.
Do not bolt on repetitive questions solely for AI visibility. Google states that normal SEO practices apply to its AI features, with no special markup requirement or guaranteed inclusion. Clear, accurate answers are worth creating for readers, not as a promise of citation.
- Can the reader apply the advice without guessing?
- Are limitations and exceptions visible?
- Does each added section resolve a distinct uncertainty?
7. Protect URLs, internal links and existing usefulness
Keep the existing URL when the page still serves the same task. A refresh rarely needs a new address merely because the year changed. Before publishing, check internal links, headings, downloadable assets, forms and mobile presentation.
Hypothetical example: a consultancy has two overlapping supplier-selection checklists. Consolidation may make sense if they serve the same audience and decision. First identify any unique material and links to each page. Preserve useful content, redirect a retired URL to the genuinely equivalent destination and update internal links.
Do not merge pages just because their keywords overlap. A beginner’s explanation and a procurement template may meet different needs. Equally, avoid removing a practical section simply because it contributes little measurable search traffic.
- Save the original page and a change log.
- Test redirects and links when consolidating.
- Change the updated date only after meaningful work; retain accurate authorship.
8. Measure reader outcomes and team effort separately
Choose success measures before publication: relevant organic clicks, completed enquiries or successful use of a resource. Record the release date and review after enough data accumulates for the page’s traffic and buying cycle. Avoid treating a single quiet week as a verdict.
Period comparisons describe what changed, not what caused it. Record promotions, seasonality, technical releases and tracking changes alongside the refresh. Unchanged pages can provide context, but they are not automatically a controlled comparison.
Track editorial efficiency separately. Hypothetical example: comparable refreshes previously required eight active staff-hours each and now require six, including review and corrections. That suggests two fewer hours per page, not proven revenue growth. Include setup time and monitor rework so faster drafting does not hide extra checking.
- Measure active hours, elapsed delivery time and revision rounds.
- Compare similar scopes and quality standards.
- Retain an inconclusive outcome when evidence is insufficient.
9. Make refreshes owned, bounded work
Give each refresh an owner, reviewer, publication checkpoint and measurement date. Keep the baseline and change brief with the task. Revisit high-change information more often than stable explanations, but avoid mandatory rewrites without a reason.
Its described workflow connects evidence-backed opportunities in DailyRise with owned tasks in RiseProjects and source-labelled period comparisons in RiseInsights. These support prioritisation and review, not guaranteed search improvement.
Connected evidence depends on provider access, permissions and source availability. WordPress, Webflow and Wix deeper access requires guided setup and permissions; do not assume automatic CMS publishing. Teams remain responsible for approving and implementing page changes. Explore the relevant products and /the-solution for the broader workflow.
- Close the task with the published URL and verification notes.
- Schedule a review rather than promising an uplift.
- Feed useful findings into the next refresh brief.
Frequently asked questions
How often should old content be refreshed?
Review according to change risk and business importance. Product instructions may need attention after a release; stable explanations may need only periodic accuracy checks.
Should I change the publication date?
Do not make an old page appear newly published. Use an accurate updated date after substantive changes, and preserve original publication information where helpful.
Should I delete pages with no organic traffic?
Not automatically. Check whether they support customers, internal navigation or a distinct business need. Archive, consolidate or remove them only after assessing that purpose.
How quickly will a refresh improve rankings?
There is no guaranteed improvement or fixed timetable. Confirm publication and indexing, then assess relevant queries over a suitable window while recording other changes.
Put this into practice with