How to Find the Words Your Customers Actually Search

Trades and their customers use different words for the same job. The trade says sanitaryware, the customer says bathroom. The trade says switchboard upgrade, the customer says my fuse box keeps tripping.

Build your site on your vocabulary and you are optimising for an audience of other tradespeople.

I spent fifteen years in plumbing and I had to unlearn this myself. The words on an invoice are precise, and precision is exactly what stops a page being found, because nobody searching for help has the vocabulary of the person who turns up to do it.

Start with what you already have

Your own enquiries. The last twenty emails, texts and voicemails. Write down how each person described the problem, in their words, before anybody translated it.

This is the most undervalued source there is. It is free, it is specific to your area, and it is actual humans rather than an estimate from a tool. Do it for a fortnight and you will have a better list than most paid research produces, because it includes the phrasings no tool records: the ones people say out loud rather than type.

Search Console. If your site has been up for a few months, Google will tell you exactly what people typed before they saw you. Open Performance, look at the Queries tab, and sort by impressions.

Two things to look for:

• Searches where you get impressions but almost no clicks. You are appearing and being passed over, which usually means you are appearing for the wrong reason.
• Searches you did not expect at all. These are the gold, because they are real demand you never thought to target.

Google's own guidance on the performance report covers how to read it properly if you have never been in there.

Then use Google itself

Free, and more honest than most tools.

Autocomplete. Start typing your service and the town and watch what it offers. Those suggestions are built from what people actually search, which is exactly the question you are asking.

People also ask. The expandable questions in the results. Each one is a question real people type, and each one is a section you could write on a page.

Related searches. The block at the bottom of the results page.

Do this on a phone, not signed in, with location set to where your customers are. Your own search history will otherwise show you what you search, which is not the same thing.

What to do with the list

Sort it into three piles.

Jobs you want more of. These get their own page. Not a line on a services page, a page.

Questions people ask before hiring. These become articles, or question headings on the service page. "What does it cost to rewire a three bedroom house" is a page, not a sentence.

Everything else. Leave it. Chasing searches for work you do not want is how you end up busy and annoyed.

That third pile is the one people struggle with, so here is the test. For each phrase, ask whether you would be pleased to get that job on a Tuesday afternoon. If the honest answer is no, it does not matter how many people search it.

One page, one job

Having found the words, the common error is to pile them all onto one page in the hope of catching everything.

A page can be about one thing convincingly. A page about emergency call-outs and bathroom renovations and commercial maintenance is a list, and a list ranks for nothing because it is not mostly about anything.

What does belong together on one page is the same job said different ways. Fuse box keeps tripping, switchboard upgrade, RCD keeps going: one job, three vocabularies, one page that uses all three naturally because a real explanation would.

Volume matters less than you think

Every tool will show you a monthly search volume and it will tempt you toward the big numbers. For a local business that is usually the wrong instinct.

A phrase searched thirty times a month in your city, where the person needs somebody that week, is worth more than one searched five thousand times nationally by people comparing prices for a job they will do themselves next spring.

What matters is intent and proximity, not volume. "Emergency electrician Papanui" is a small number and nearly all of it is work.

There is also a timing dimension nobody mentions. The same person searches three different ways across a fortnight: first the symptom, then the fix, then somebody to do it. You want the page that catches the last one, and ideally the first one too, so that when they get to the third they already know your name.

The mistake almost everyone makes

Picking the phrase that describes the business rather than the job.

Nobody searches for "quality workmanship" or "trusted local tradesman". They search for the thing that is broken, or the thing they want installed, usually with a place attached. The page needs to be about that thing, named the way they name it.

That is most of what getting found on Google comes down to, and it is why the first conversation is about what you want more of rather than about keywords at all. Once you have the list, whether to aim at the city or the suburb is the next decision, and it changes what you do with every one of those phrases.