The local SEO checklist for trades and service businesses

Do them in this order: Google Business Profile, then reviews, then one real page per service and per town, then a small handful of UK citations, then content. That sequence is not arbitrary. It is roughly the order of return per hour spent, and most trades I speak to have worked through it backwards, paying for a directory blast and a neglected blog while the primary category on their Google listing is still set to the wrong thing.

Below is the checklist I actually work through, with the honest version of how long each item takes to show up. If you only ever finish the first two sections, you will collect the majority of what is available to you. None of it needs a budget beyond your own time, apart from one paid category I will flag as usually not worth it.

Google Business Profile comes before the website

For searches like "plumber near me" or "electrician maidstone", the map pack sits above the ordinary blue links and takes most of the clicks. That block is fed mainly by your Business Profile, not by your site. Google weighs three things: how close you are, how relevant you look, and how prominent you are.

You cannot move your unit and you cannot invent prominence overnight. Relevance you can fix this afternoon, for free. That is why this section is first.

The primary category is the single highest-leverage field

You get one primary category and up to nine additional ones. The primary is by far the strongest signal for which searches you turn up in, and it is the field I find wrong most often. An electrician sitting on "Electrical repair shop" instead of "Electrician". A builder on "Contractor" instead of "General contractor" or, better, "Home builder". A boiler engineer on "Plumber" when half his money is in heating.

Check it in ten minutes. Search your main money term in a private window, look at the top three map results, and read the category printed under each business name. If all three say Electrician and yours says something else, change yours. Expect a week of wobble afterwards, so do it once and deliberately rather than experimenting every fortnight.

Secondary categories: only the work you want

Each additional category opens up another set of searches, so it is tempting to add ten. Don't. Wrong categories dilute what Google thinks you are and bring in enquiries you have to turn down. A gas engineer who adds "Air conditioning repair service" without doing air conditioning has bought himself phone calls with no revenue at the end of them.

Four or five true ones is plenty. If you would happily quote for it tomorrow, it belongs there.

Fill in every service, with a description on each

The services section is free space that most profiles leave half empty. Each service takes a name, either from Google's list or one you type yourself, plus a description of up to 300 characters. Google matches that text against what people type.

Write them like a person, not a keyword list. "Full and partial rewires for older properties, including consumer unit upgrades and certification. Most three-bed rewires take four to five days." That is more useful than "rewire electrician rewiring services". Mention your town once where it reads naturally and leave it there.

Service area or address: be honest, because the alternative bites

If customers cannot turn up and be served at your premises, remove the address and set a service area instead. A pretend shopfront at your home address is one of the quickest ways to get a profile suspended, and getting it reinstated takes weeks you cannot afford.

Now the part nobody selling SEO wants to say: your service area does not make you rank across it. You rank in a radius around your pin. Listing twenty towns does not place you in twenty map packs. In practice you will be strong in your own town, competitive within five to eight miles, and thin beyond ten to fifteen unless the area is rural and there is barely anyone to compete with. If most of the work you want is twenty-five miles away, no amount of on-site tinkering fixes that. A second base or paid advertising will.

Hours, and the bank holidays everyone forgets

People filter by "open now" more than you would think, and wrong hours quietly cost you calls. Set the standard week accurately, then add special hours for the days that catch everyone out: Christmas and Boxing Day, New Year, both May bank holidays, Easter and the late August weekend.

If you genuinely take emergency calls at two in the morning, say so and answer the phone. If you don't, don't claim it, because the review that follows an unanswered emergency call is expensive.

Photos: a trickle beats a dump

Forty photos uploaded in one afternoon does less than three or four a month, every month. Add real work: the van, the finished bathroom, the consumer unit before and after, the roof at the point it stopped leaking. Stock photography reads as stock photography to customers, and it tells Google nothing about you.

Geotagging your images before upload has no effect I have ever been able to measure, and Google strips most of that data anyway. Skip it. Do keep the cover photo and logo current, and do make sure the pictures are sharp, because a blurry photo of a good job loses you the enquiry.

Products, Q&A and Posts, in that order of usefulness

Products are shelf space you are not paying for, and they work well for anything you can put a fixed price on. Boiler service, £95. EICR from £180. Gutter clear on a three-bed semi, £70. Prices on the profile pre-qualify people before the phone rings.

The Q&A section is public and you are allowed to ask and answer your own questions, which almost nobody does. Seed the six you get asked every week: do you charge for quotes, do you cover my postcode, are you Gas Safe registered, how soon can you come out, do you take card, is there a call-out fee. Otherwise a competitor or a chancer answers for you.

Posts are last for a reason. They help a little with conversion and, in my experience, do very little for rankings. Post when there is something real to say. Building a weekly ritual around them is not a good use of your evenings. There is more detail on all of this in my guide to the Business Profile itself.

Reviews: the biggest lever you actually control

Prominence, the third of Google's three factors, is largely reviews. Count matters, but it is not the whole story. Recency and rate carry a surprising amount of weight, which is good news if you are starting from four reviews and your competitor has ninety from 2021.

Ask at the moment of relief, not three days later

Ask in person, on the drive, while the customer is still pleased the leak has stopped. Then send the link by text before you pull away. Grab the short link from your profile under "Ask for reviews" and save it as a text template on your phone so it takes six seconds.

Ask face to face, then follow it with a text. On that method a third to a half of customers leaving a review is a fair target. Emailed a week later, expect nearer one in twenty. Treat both as targets rather than measured results. The point is the size of the gap between the two methods, not the exact figures.

Never filter, never pay. Asking only the happy customers, or offering a discount for a five-star, breaks Google's rules and the platform is reasonably good at spotting it. It also reads as fake to the people you are trying to convince.

Velocity beats a big old total

A hundred and twenty reviews with the most recent one two years old performs worse, and reads worse, than forty with four from this month. For a one-van trade, two to four a month is a realistic and defensible target. That is one in every ten jobs, which is what asking properly gets you.

Do not try to catch up in a week. Twenty reviews in seven days after two silent years is the exact shape Google's filters look for, and some will simply vanish.

Reply to every one, briefly

A sentence is enough, within a couple of days, mentioning what the job was and where. It gives Google clean text and it shows prospects you are still trading. For a bad review, stick to facts, offer to put it right, sign your name and stop. Prospects read your replies more carefully than they read the reviews, and a calm reply to an unfair one-star does more good than the review does harm.

The website layer

A real page per service, and per town

The paragraph that reads "we cover Maidstone, Ashford, Canterbury, Faversham, Sittingbourne and Whitstable" does nothing at all. There is no content on that page about any of those places, so there is nothing for Google to rank and nothing for a reader to trust. One page cannot be the best result for six towns.

A town page earns its place when it contains four hundred words that could only have been written about that town. Jobs you have done there and roughly what they cost. The housing stock, because Victorian terraces with lead supply pipes are a different job from a 2015 estate. How long it takes you to get there. A review from a customer in that postcode. Photographs from that street.

The test is simple: if you cannot write it without a template, do not publish it. Ten thin near-identical pages are worse than none, because they teach Google that your site is padding. Do services first, since intent is higher and competition per page is lower, then add the two or three towns you actually want more work from. That is how I structure a site when I build one: a page per real search, rather than a list of place names in a paragraph.

NAP consistency, decided once

Name, address, phone. Pick one canonical version, write it in a note, and copy-paste it everywhere for the rest of the business's life. If the limited company name differs from the name on the van, use the trading name people recognise and keep it byte-identical across every listing.

One phone number, one format. Call-tracking numbers that differ from the one on your Google profile are a real risk of confusing the match between your site and your listing. If you need tracking, use dynamic number insertion so crawlers and Google see your genuine number.

LocalBusiness schema, and only the fields that earn their keep

Use the most specific type available rather than the generic one: Plumber, Electrician, RoofingContractor, HVACBusiness, Dentist. Then fill in name, url, telephone, image, address with the full postcode, geo coordinates, areaServed, openingHoursSpecification, and sameAs pointing at your Google Business Profile and your Facebook page. That is the useful set.

Leave aggregateRating alone unless the reviews are genuinely collected and displayed on your own site. Marking up your Google stars as if they were yours is against the guidelines and it is the sort of thing that gets rich results pulled. Schema does not rank you. It removes ambiguity about who and where you are, and it takes half an hour once.

Link from the service page to the town page, in the text

A sentence in the body of the rewiring page saying that most of my rewires are in a particular town, with that phrase linking to the town page, is worth more than a footer stuffed with thirty place names. The footer block is close to worthless and looks like what it is. Link back up from the town page to the service page too, so the pair reinforce each other.

The map embed and the phone number

Embed the map from your own Business Profile, using Share then Embed, rather than dropping a generic pin. I have never measured a ranking effect from it. What it does is reassure a stranger that you are a real business at a real place, which is most of the job on a contact page.

Add loading="lazy" to that iframe. An embedded map is often the single heaviest thing on a small trade site and it will happily wreck your mobile load time if you let it, which matters more than the embed itself. I have written about that trade-off in why website speed costs you customers. Put your phone number in real text inside a tel: link, in the header and the footer, never as part of an image.

Citations: about six, then stop

The £99 "submit your business to 200 directories" package is dead money. It has been for years. What still helps is a small number of places Google clearly uses, filled in consistently.

• Bing Places and Apple Business Connect. Both free, both take fifteen minutes, both send you the occasional call you would otherwise never have received.
• A Facebook page with the same name, address and number, even if you never post to it.
• Yell. The free listing is worth claiming. The paid packages are difficult to justify on the numbers I have seen.
• Thomson Local. Five minutes, low value now, harmless.
• Your Companies House record, matching the address on your site.
• Trade bodies, which are the ones that genuinely count: Gas Safe Register, NICEIC or NAPIT, the ECA, the FMB, TrustMark, Which? Trusted Traders. Real links from sites customers already trust, and half your prospects will check the register before they ring.

Checkatrade, MyBuilder and Rated People sit in a different category. They are lead generation, not SEO, and they should be judged only on cost per won job. Fees vary a lot by trade and area and they change, so check current pricing yourself rather than trusting a figure in a blog post. What I can tell you is that the outcome splits sharply: some trades win steady work from them for years, others spend four figures and get three tyre-kickers. Track it for three months and cancel without sentiment if the maths does not hold.

Content: fewer pages, better ones

Write down what people ask you on the phone

Keep a note on your phone for a fortnight. Every question that comes up twice is a page. "How much does a rewire cost." "Do I need to be there for the EICR." "How long does a boiler swap take." "What is a landlord gas safety certificate and how often do I need one." These are the searches, in the customer's own words, and you already know the answers better than anyone writing about them.

Search Console gives you the same thing for free. Performance, then Queries, then look for terms where you sit somewhere between position eight and twenty. Those are things Google already half-associates with you, and a proper page usually moves them.

Plan seasonal pages three months early

Boiler servicing demand climbs through September and October, so that page needs publishing in July. Gutters and roofing spike after the first autumn gales. Drainage calls follow heavy rain. Garages get their MOT peaks in March and September because of the registration cycle. A page published the week the demand arrives is late, because Google needs a few weeks of trust before it will rank anything new.

One good page a year beats twenty thin ones

"What a full rewire costs in Kent in 2026", written from your own last twelve invoices, with ranges by property size and an honest account of what pushes a price up, will out-earn everything else on your site. It gets shared, it gets linked, and it answers the question better than any national content mill can, because they do not have your invoices.

Twenty three-hundred-word posts do nothing. And if you hate writing, genuinely do not force it. A well-built site with five honest service pages and a working review habit outperforms a good site with an abandoned blog. If your problem is that nothing ranks at all, the cause is usually structural rather than a shortage of posts, which I have picked apart in why your website is not showing up on Google.

Measuring it without kidding yourself

Two free tools cover almost everything. In your Business Profile, open Performance: calls, messages, direction requests, website clicks, and the breakdown of the actual search terms people used. Watch calls and direction requests above everything, because they sit closest to money. Google only keeps six months of it, so export a copy each month or you will lose your own baseline.

Search Console handles the site. Compare the last three months against the previous three, filter by page, and look at impressions, clicks and average position together. Rising impressions with flat clicks usually means you are appearing for the wrong things, which is a copy problem, not a technical one.

What not to obsess over is your own rank check from your own postcode. Local results shift street by street and by search history. Use a private window, accept the fuzziness, and judge the trend rather than the reading.

Realistic timings, since this is where expectations break: a category or services fix often shows in three to fourteen days. A review habit takes three to eight weeks before it registers in the pack. A new service page needs four to eight weeks to settle, sometimes twelve. A town page takes eight to twelve, and may never rank if you are simply too far away. Citations have no visible date, ever. Do them once and forget them.

Three questions I get every time

Do I need a separate page for every town I cover?

No. Only for towns you genuinely work in and can write about honestly. Two or three real town pages beat fifteen templated ones, and the templated ones can hold the whole site back.

How long before local SEO pays for itself?

Business Profile work often inside a month. The website layer, three to six months. If you need work booked next week, that is a job for paid ads, and I will tell you so rather than sell you a retainer.

Are paid directories worth it?

As lead generation, sometimes. As SEO, almost never. Judge them on cost per won job over a three-month test, then keep or cancel on the number.

If you do nothing else from this list, fix your primary category and start asking for reviews on the drive. That is an afternoon's work and it is worth more than everything below it put together.

If you want a second opinion on where your own listing and site are losing work, send me the details and I will go through it. If there is nothing worth paying for, I will say that too. Otherwise, how I approach SEO explains what the work looks like and what it costs.