Search Console shows which local searches Google already connects with your site, and which pages it picks. That is a better starting point than a keyword list.
Quick answer
Export Query by Page data, filter the queries for place names, group them by town and by type of intent, then check which page is serving each group and at what position. That shows where there is demand you already appear for, where the wrong page is ranking, and which locations deserve a page or more support.
Step 1: export Query by Page data
In Performance, set a date range of at least three months, add the Queries and Pages dimensions, and export. Local analysis needs both: the query tells you what people asked, and the page tells you what Google chose.
Step 2: filter and group by place
Filter queries with a regular expression containing the places you care about, such as `southampton|winchester|romsey|hampshire`. Then total the impressions, clicks and average position per place. Weight position by impressions, because a handful of impressions at position 5 should not outweigh thousands at position 80.
What it showed on my site
I ran this on the Query by Page export for peacocksearch.co.uk, covering roughly July to October 2026. Grouped by place, queries containing each name gave:
- Southampton: about 13,700 impressions across 58 queries, weighted average position around 65, no clicks
- Hampshire: about 9,700 impressions across 118 queries, average position around 77
- Chandlers Ford: about 770 impressions across 7 queries, average position around 72
- Salisbury: about 480 impressions, average position around 60
- Winchester: about 420 impressions across 34 queries
- Gosport and Havant: about 330 impressions each
- Fareham: about 170 impressions
- Romsey: about 70 impressions, average position around 22
- Portsmouth: about 50 impressions
A few things follow.
Demand is not where the base is
Romsey is my base, and it ranks well (around position 22 on average) but has very little search demand. Southampton and the county-wide terms carry nearly all the impressions. A local business's home town is often not its biggest market.
Intent decides where you rank
For Southampton, queries that ask for an agency, company or services ("seo services southampton", "seo agency southampton") sit at position 80 or lower. Queries that ask for a consultant or expert ("seo consultant southampton", "seo consultants southampton", "seo expert southampton", "seo consultant in southampton") sit around positions 25 to 30. The page matches the consultant intent. It does not match the agency intent. That is a content and positioning decision, not an optimisation to squeeze out.
The freelance angle is closest to page one
"Freelance seo hampshire" has about 260 impressions at an average position of 13.4, and "freelance seo consultant southampton" averages 10.5. Those are the closest queries to page one in the data, although on small volumes, and they match what I actually offer.
The wrong page is sometimes picked
For "seo consultant winchester", the Hampshire page appeared at around position 33 and the Winchester page at around 19. For "seo chandlers ford", the Hampshire page and the Southampton page split the impressions about evenly, and no page exists for the town. These are cases where internal linking, titles and page purpose matter more than new pages.
Towns without pages still show demand
Chandlers Ford, Gosport, Havant and Fareham all show impressions without a dedicated page. That does not mean each needs one. It means they are worth watching, and that a section on a hub page might be enough, as discussed in how to build location pages that do not feel like doorway pages.
Step 3: turn it into decisions
- High impressions, poor position, right intent: strengthen the page, internal links and proof
- High impressions, wrong intent: decide whether to compete or to target the queries you suit
- Close to page one: prioritise these. Small improvements can move them to where clicks happen
- Wrong page ranking: adjust titles, headings and internal links so the intended page is the best answer
- Demand without a page: check the evidence and whether you can say something distinct, before creating one
Step 4: re-check
Make the changes, note the date, and compare the same groups after a few weeks. This is the same method I used to decide what to change on the Hampshire and Southampton pages. I will re-check in about four weeks.
Common mistakes
- Looking at clicks only, when impressions and position tell you where you stand
- Averaging position without weighting by impressions
- Creating a page for every place that appears
- Ignoring query intent
- Not checking which page is ranking
When to get support
The Visibility Review includes a Search Console review like this, and local SEO support covers acting on it. See also why local pages get impressions but no enquiries.
FAQ
How much data do I need?
At least three months, and more for seasonal businesses. Small samples are noisy, particularly for position.
Can I do this without exporting data?
Yes, using the Performance filters, but an export lets you group and total by place far more easily.
Do impressions at position 80 matter?
Not for clicks, but they show the topics Google already associates with your site, and which pages are being considered.