Which Local Searches Are Actually Worth Chasing (And Which Aren't)

Ask a plumber which search they'd most like to rank for and you'll usually hear "plumber" or their town on its own. Big word, big number, almost no phone calls in it.

Most owners pick targets by sorting a keyword tool by volume and taking from the top. That's close to the worst possible method, because volume and buying intent pull in opposite directions. The searches that fill a diary are small and unglamorous.

Here's how to tell them apart, in an afternoon, with a pencil.

Buyer intent and research intent produce two different pages

This used to be a judgement call. It isn't any more, because the results page looks visibly different depending on what someone typed.

Whitespark hand-ran 540 local queries across six industries and three US cities, published 12 May 2025. Queries of the "primary care clinic Phoenix" shape returned a local pack 93% of the time and an AI Overview 15%. Informational ones like "how long does an eye exam take in Phoenix" flipped it: AI Overview 92%, local pack 6%.

American data, and over a year old, so check it on your own phone. The pattern holds. Where the map pack shows, the AI answer recedes. Where the search is a question, Google answers it and nobody scrolls.

Heating is the clearest British example. "Boiler upgrade scheme" and "heat pump grant" carry real volume, but the person typing them wants the government's figure (£7,500 towards an air source heat pump in England and Wales). GOV.UK answers that, and so does the AI box. "Air source heat pump installer [your town]" has a fraction of the volume and a buyer behind it.

"Near me" is not something you can put on a page

Owners ask for this constantly and it isn't something you can optimise. Google's local ranking guidance names three factors: relevance, distance and prominence. "Near me" is resolved by the searcher's device location against distance, so putting the phrase in a page title achieves nothing.

From the same page, word for word: "There's no way to request or pay for a better local ranking on Google." Reviews do get a mention though. Google says more of them, and better ratings, can help.

Service plus town works. Forty copies of it don't.

A page for a town you genuinely cover earns its keep, but only if it reads like you wrote it. Name the estates you get called to. Say how long the van takes from your yard on a weekday morning. Mention the housing stock, because a road of 1930s semis with back boilers is a different job to a new build estate with combis in the loft.

Spinning one up for every village within fifty miles is a different thing entirely. Google's spam policies list "Doorway abuse", pages "created to rank for specific, similar search queries", and give multiple pages targeted at cities as the example. Scaled content abuse sits alongside it, explicitly covering generative AI used to produce pages without adding value.

Our rule of thumb: if you'd be embarrassed to send the page to a customer, don't publish it.

Every free tool hides the searches you live on

Keyword Planner is where most owners start, and the small print matters. It's a twelve month average, historical stats only cover exact matches, and the figures are rounded, so volumes across locations "might not add up as you'd expect". Without meaningful ad spend you get wide bands instead of numbers. It groups close variants into one figure too, so "plumber Bristol", "Bristol plumber" and "plumbers in Bristol" often show the same volume. Add those together and you've counted the same demand three times.

Ahrefs compared 72,635 keywords against Search Console data and found Keyword Planner came in higher than the real figure in 91.45% of cases. Trends won't rescue you either. It's scaled 0 to 100, it "only shows data for popular terms, so search terms with low volume appear as 0", and Google adds statistical noise deliberately. A flat zero for "emergency electrician Chippenham" is a floor effect, not an absence of customers.

Your own data is better. Business Profile, then Performance, then Searches, last six months, gives you the phrases people typed when your profile appeared. Anything under fifteen searches in a month shows as "<15", and most of a single town trade's list reads like that. Normal, not a fault, and usually the ones that convert. They're impressions, not enquiries, so don't count them as leads.

Search Console holds sixteen rolling months and caps the table at 1,000 rows. Sort by clicks, not impressions. Low volume queries get stripped out for privacy so the rows never add up, and Google's own data anomalies log records a logging error that stopped impressions being reported accurately from 13 May 2025 until 27 April 2026, never backfilled. Any impression comparison across that window is worthless.

One more. Search Console's generative AI report shows impressions only, with no query dimension at all, so nobody can do keyword research on AI Overviews yet. Anyone selling you that is guessing.

How to spot a term you'll never rank for

Ignore difficulty scores. Open the search on a phone with location on and count how many of the top ten are businesses like yours, and how many are Checkatrade, MyBuilder, Rated People, Bark, Yell or a national brand. If it's mostly the second lot, park it.

Then there's age. Ahrefs' study, updated May 2025, found only 1.74% of newly published pages reach the top ten within a year, and that 72.9% of those in the top ten now are over three years old. You can still win. Not in eight weeks, and not against a long established directory on a national phrase.

The afternoon method

Pencil, envelope, phone, about two hours.

• Twenty minutes. Business Profile, Performance, Searches, last six months. Write down every phrase that isn't your business name, including the "<15" ones.
• Twenty minutes. Search Console, United Kingdom, last three months, sorted by clicks. Add anything new.
• Fifteen minutes. Type your services plus your town into Google and harvest autocomplete, People Also Ask and Related Searches. Live, free, nothing hidden behind a threshold.
• Thirty minutes. Run the list through Keyword Planner, location set to your nearest targetable city. Expect bands, never add variants together, sort into rough tiers.
• Thirty minutes, and this is the step that decides it. Search each candidate on a phone with location on. A local pack near the top means chase it. An AI Overview and no pack means research intent, so expect reads not calls. Directories and national brands, leave it alone.
• Five minutes. Seasonality in Trends, UK, five year view. Ignore the zeros.
• Ten minutes. Do the sum. Twelve searches a month, [your realistic enquiry rate] of those getting in touch, [your average job value] each. That tells you more than any five figure national volume.

A word on the numbers you'll be quoted

If someone quotes "76% of people who search for something nearby visit a business within a day", ask for the date. It's a Think with Google figure from around 2016. "46% of all Google searches have local intent" traces back to a Google representative at a conference in 2018. Check the date and the country before a number changes your plan.

One is worth keeping. Google has said for years that roughly 15% of the searches it sees each day are ones it has never seen before. Your best converting phrase might not exist in any tool.

If you'd rather someone did the thorough version and showed you the working, that's a fair chunk of what we do. The envelope version is free though, and it beats a guess.

How many searches a month does a term need before it's worth targeting?

There's no universal threshold, because it depends on what a job is worth to you. A phrase showing "<15" in your Business Profile can be excellent for a heating engineer at [your average job value] and pointless for a salon at [your average treatment price]. Do the arithmetic with your own numbers.

Can I check demand for a specific postcode like BS7?

Not in Keyword Planner. Its volume data resolves to city level, so a postcode district won't return its own figure even though plenty of people search that way. Your Business Profile search terms and autocomplete are the closest you'll get.