SEO
Find the pages that are nearly on page one
Use Search Console to find near-page-one SEO opportunities, diagnose the right improvements and measure progress without confusing average position with certainty.
OppRise Editorial · Reviewed · 8 min read
The short answer
Find near-page-one opportunities by reviewing query–URL pairs in Google Search Console, initially around average positions 8–20. Validate intent, consistency and business relevance, diagnose a specific fix, then measure comparable windows. Average position is a screening signal, not a guaranteed route to page one.
What to take away
- Prioritise relevant query–URL pairs rather than sitewide averages.
- Check intent, technical constraints and data stability before rewriting.
- Rank concrete improvements by relevance, confidence, effort and dependencies.
- Measure search outcomes and team efficiency separately, without claiming causation.
1. Treat near page one as a shortlist, not a promise
A page averaging position 12 is not necessarily sitting two places below a stable first page. Search results vary by query, device, location and result layout. Google’s Search Console documentation explains that average position reflects the topmost result from your property or page, depending on aggregation. It is not a fixed rank.
Use positions 8–20 as an initial working range, not an SEO rule. Include pages already inside that range’s upper end because weak clicks may reveal an opportunity. Exclude pages that cannot serve the searcher or support your business.
All examples below are hypothetical. A commercial refrigeration repair page averaging position 14 for an urgent service query might deserve attention before a position-11 article about refrigerator history. Proximity matters less than relevance, demand and a fix your team can actually deliver.
- Define the commercial outcome: qualified enquiries, purchases or another useful action.
- Choose one market and search type before comparing candidates.
- Record why each shortlisted page deserves work beyond its position.
2. Build query-and-page evidence in Search Console
Open the Search results performance report. Select Web search, your target country and a consistent date window. Start with 28 complete days; extend the window for sparse data, accepting that a longer window can hide recent changes. Enable clicks, impressions, CTR and average position.
Review queries, then select a candidate query and inspect the Pages tab. Reverse the process for a promising URL: select the page and inspect its queries. Export the relevant filtered views and shortlist query–URL combinations in your spreadsheet. Do not assume separate query and page exports form a joined dataset.
For a Perth bookkeeping firm, separate 'bookkeeper Perth' from 'bookkeeping course Perth'. They may appear against the same page, but only one fits the service. Search Console omits some queries for privacy, and aggregation affects totals, so visible query rows are not a complete demand inventory.
- Keep query, URL, country, device, dates, impressions, clicks, CTR and position together.
- Record filters alongside exports so another person can reproduce the shortlist.
- Compare query–URL pairs, not a sitewide average position.
3. Separate persistent opportunities from noisy averages
Inspect the candidate across shorter intervals and device segments. An average of 13 could describe a reasonably stable result, or a mixture of strong desktop visibility and weak mobile visibility. Those situations require different investigations.
Suppose a holiday accommodation page averages position 15 across eight weeks but has recently moved much higher. Rewriting it immediately could interrupt useful observation. Conversely, one short burst of impressions for an event that has finished may not justify ongoing work.
Compare equivalent windows and annotate promotions, outages, migrations and seasonal events. There is no universal minimum impression count. With thin data, lower your confidence and favour small, reversible improvements rather than treating uncertain estimates as precise priorities.
- Check whether the same query and URL recur across multiple intervals.
- Separate mobile and desktop when the combined average hides material differences.
- Mark each candidate as stable, changing or insufficient evidence.
4. Check whether the page deserves to answer the query
Review current results for a small set of important queries in the intended market. Note the dominant page types and needs they address. A manual search is a contextual snapshot, not proof of what every customer sees.
For 'commercial dishwasher dimensions', a product page may need verified measurements, installation clearances and a specification sheet. A broad article about choosing kitchen equipment may be the wrong destination, even if its average position looks promising.
Distinguish a missing answer from a mismatched page. Improve the existing page when its purpose fits. Consider a separate resource when the query requires a genuinely different experience. Do not force buying advice, troubleshooting and transactional content into one unfocused page simply to collect more keywords.
- Can the visitor complete the task implied by the query?
- Is the appropriate page type a service page, category, product or guide?
- Can your team supply reliable information that is currently missing?
5. Diagnose the constraint before commissioning copy
Check that the intended URL is indexable and that canonicalisation and redirects match your intended structure. If Search Console attributes performance to another canonical URL, investigate before briefing an editor to improve the wrong page.
Next inspect content accuracy, navigation and competing URLs. Two pages appearing for similar searches do not automatically prove harmful cannibalisation. Determine whether they serve distinct needs, whether the wrong destination repeatedly appears and whether their overlap confuses users.
A Sydney electrician might have a general service page and a dedicated switchboard-upgrade page. Clarifying their scopes and linking between them may be more appropriate than merging them. Consolidation can be valuable, but it also creates redirect, content-loss and implementation risks. Do not infer backlink weakness from a position report.
- Technical problem: assign investigation before editorial expansion.
- Wrong destination: review page scope, internal links and overlapping content.
- Missing substance: identify the exact question and evidence required.
6. Rank a specific fix, not an entire URL
Give each candidate a proposed intervention, business relevance, evidence confidence, effort estimate and dependency list. This is more defensible than multiplying impressions by an invented click-through-rate uplift.
A workwear category missing accurate waterproofing distinctions might need a merchandiser’s review and a short comparison block. A legal services guide requiring extensive practitioner review could have greater potential but substantially higher delivery cost. Both can be worthwhile; they should not occupy the same quick-fix queue.
Use a simple high, medium or low assessment rather than false numerical precision. Prioritise relevant demand, a clear information gap and a feasible fix. Park work where the evidence is weak or the required expert is unavailable.
- Write the hypothesis: 'Adding verified clearance dimensions will better answer installation queries.'
- Estimate research, drafting, approval, publishing and quality-assurance time.
- Name an owner and any expert or developer dependency.
- Define the measurement window before approving work.
7. Improve usefulness and discovery together
Google’s people-first content guidance favours useful, reliable material over content created primarily to attract rankings. Add what helps the visitor decide or act, not a predetermined word count.
A roof-restoration page might need a clear inspection process, exclusions and genuine project evidence. Its title should accurately describe the service and location without inventing prices, credentials or availability. Review headings and introductory copy for clarity rather than repeating every query variation.
Add relevant internal links from pages where a reader would naturally need the destination. A maintenance guide could link to the restoration service when explaining when repairs require professional assessment. Avoid inserting the same keyword-rich link throughout unrelated articles. Package changes around one hypothesis where practical, and document bundled changes when separating them would make the page worse.
- Verify new facts with the responsible subject-matter expert.
- Keep useful existing content unless there is a clear reason to remove it.
- Check links, mobile presentation and enquiry or purchase paths after publishing.
8. Measure movement without claiming causation
Save the baseline, publication date and change log. After allowing time for discovery and fresh reporting, review the same query–URL combinations with matching filters. Track impressions, clicks, CTR and position together, then examine relevant organic landing-page outcomes separately.
A page can gain impressions and clicks while its average position worsens because it starts appearing for additional queries. Equally, improved CTR with falling impressions may not mean greater business value. Avoid reducing the review to one green arrow.
Period comparisons show what changed, not what caused it. Demand, competitors and other website changes can influence results. Similar unchanged pages can provide context, but are not a randomised control. Set a review date and decide whether to retain, refine or investigate rather than repeatedly editing in response to daily fluctuations.
- Separate the original target-query group from newly appearing queries.
- Report downstream outcomes with their measurement source and limitations.
- Record inconclusive results instead of forcing a success narrative.
9. Reduce handoff time while keeping human judgement
Use one task template containing the filtered evidence, diagnosis, proposed change, owner, approval requirement and review date. Measure analyst preparation time, clarification rounds and rework as well as delivery time. Compare similar batches and include checking overhead before claiming a saving.
Hypothetically, reducing preparation from 50 to 35 minutes saves 15 minutes per comparable brief. That is a workflow saving, not evidence of better rankings.
Its product scope includes near-page-one prioritisation in RiseVisibility, evidence-backed opportunities in DailyRise and assigned delivery in RiseProjects. RiseInsights supports source-labelled period comparisons, not causal proof. Connected evidence depends on provider access, permissions and data availability. Website changes still require your team’s review and implementation; these products do not guarantee inclusion or rankings.
- Review the relevant product pages and /the-solution before planning a workflow.
- Keep approval and publishing responsibilities explicit.
- Measure time saved only after allowing for verification and rework.
Frequently asked questions
Does position 11 mean my page is on page two?
Not reliably. Search Console reports an average, and result layouts vary. Treat position 11 as a discovery signal rather than a fixed location.
Should I improve every page ranking between 8 and 20?
No. Exclude irrelevant queries, unsuitable destinations and work without a clear user benefit. Commercial fit and a feasible improvement matter more than proximity alone.
How long should I wait before reviewing a change?
Allow for discovery and sufficient reporting data. Use several weeks as a practical review starting point, extending for low-volume or seasonal pages, not as a ranking deadline.
Should I create a new page for every promising query?
No. Group queries by user need. Improve an existing suitable page; create another only when the need warrants a distinct destination.
Put this into practice with