Finding local search terms that bring work

The translation happens in every business, it is simply never written down. Whoever answers the phone in the office hears it is leaking under my sink and enters leak at the trap, with the address, into the calendar. That translation takes half a second and is so routine that nobody notices it any more. The original sentence — precisely the wording that would also have been typed into a search box — disappears without trace.

Then there is the division of labour around text. Website content is usually supplied by the owner or the master craftsperson, that is, by the people with the most pronounced technical vocabulary in the business. Those who hear the customer sentences verbatim sit at the phone or stand at the counter, and are often not asked at all when the texts are written. The knowledge exists; it is simply in the wrong place.

The effort is modest: a sheet beside the phone or a shared note on the computer, plus the rule that the caller's first sentence is written down exactly as it was spoken. Not the diagnosis, not the job, not the calendar entry — the wording. Alongside it three details that make later sorting possible: which place the call came from, how urgent it was, and whether it turned into work.

After a few weeks a business with a normal volume of calls has enough lines to see patterns (project experience). The evaluation needs no technology: group identical phrasings, sort the groups by frequency, and note beside each one which page of the website would answer that question today. In most businesses that last column stays conspicuously empty.

The second insight from the list concerns urgency. Calls where somebody is standing next to the problem and waiting require a different page from calls where a bathroom renovation is being considered in spring. Both are valuable, but they need different entry points — and usually different pages.

Only now does the search box come into play. Typing in your own terms produces completion suggestions, and related queries appear at the foot of the results page. Neither is an official figure; both are indications of which word combinations are actually common. For a trade business that is enough, because the point is not forecasting but the question of whether your own vocabulary matches demand.

A fixed procedure helps: type the problem, then the problem with the place, then the problem with cost, then with near me. Those four variants cover the usual intents. Whatever appears more than once goes on the list; whatever shows up only once stays out for now. The list does not need to be complete, it needs to be workable.

Two traps are worth avoiding. The first is fascination with big terms: bathroom renovation sounds like plenty of demand, but it is contested nationwide and says nothing about intent. The second is your own search history, which distorts the suggestions — typing your own terms daily means being offered them preferentially and mistaking that for demand. A window without a signed-in account gives the soberer picture.

As soon as the list of terms exists, an obvious temptation appears: one page per town, text written once, town name swapped, twenty pages done. The approach is old, it is quick, and it usually no longer works. It fails with the reader first, long before any search engine passes judgement.

A searcher landing on such a page notices within seconds that the town name is the only locally specific thing about it. What is missing is exactly what matters: how long does the business take to get here? Does it also come to the part of the village beyond the level crossing? Has it worked here before? Who will I reach if I call? A page that answers none of those four questions is not a location page, it is a placeholder with a town name in it.

A sound location page contains travel time and route, the outlying parts that belong to the area, at least one real project or a typical construction style from that place, the named responsibility inside the business, and the hours when somebody can be reached. That cannot be generated in twenty variants from a template — which is exactly why it has value. How it interlocks with the business profile in map search is a topic of its own; page and profile have to carry the same details.

Twenty collected phrasings do not become twenty pages. The cut comes from bundling: all terms sharing the same intent and the same route to a solution belong on one page, even when they are worded differently. Dripping tap, leaking mixer and water running under the sink describe the same job and need no three pages, only one that picks up all three phrasings in its text.

The boundary runs where the intent or the sequence changes. Bathroom renovation cost and dripping tap do not belong on the same page, even though both sit in the same trade. As a rule of thumb: if the next sensible step for the visitor is a different one, a separate page is needed. Anybody unsure about the cut is welcome to raise it with us through contact.

The most reliable signal comes not from an analysis but from the phone. When callers start using the wording of the page — I read that you also come out on Saturdays — the page has been found and read. Those sentences are valuable and are nevertheless rarely written down. They belong in the same list the collection started with.

The second signal is the composition of enquiries. Are more enquiries coming from the places for which pages with real substance were written? Have the enquiries become more concrete because the page anticipated the follow-up questions? Is the share of enquiries outside the range of services falling? These are observations rather than measurements, but they are immediately available inside the business and usually unambiguous.

And finally the uncomfortable point: if nothing has changed after three months, then either the intent was assigned wrongly, the next step on the page is unclear, or the page still answers the question the business has rather than the question the searcher has. In all three cases more text will not help; a fresh look at the list will — and, in case of doubt, a conversation in which somebody from outside reads the page without knowing the trade. We are happy to go through that with you via contact.