A local business website has one job that matters and a lot of jobs that feel like they matter. The one that matters is this: somebody who is already fairly likely to hire you lands on it, and leaves either convinced or gone. Almost everything else — the color scheme, the stock photography, the number of pages — is downstream of that moment.
That framing is useful because it tells you what to fix first. If your site is losing people, it is almost never losing them because it is ugly. It is losing them because it is slow, because it does not answer the question they arrived with, or because the next step is not obvious enough to take with a thumb while standing in a driveway.
Job one: load before they give up
Speed is the least glamorous item on this list and the one with the most direct effect. A visitor who taps a search result on a phone is on a mobile connection, often a mediocre one, and they have three or four other results still sitting behind them. Every second of blank screen is a chance to go back.
The usual culprits on a local business site are boringly consistent, and you can check most of them yourself:
- Hero images uploaded straight off a phone or a camera at three or four thousand pixels wide, served at full size to a 390-pixel screen. This one item is frequently more than half the weight of the whole page.
- A slider or carousel on the homepage, which loads every slide even though most visitors see only the first one, and which almost nobody clicks through.
- Third-party scripts: a chat widget, a review badge, a booking embed, two analytics tools and a tracking pixel, each of which was added for a reason and none of which was ever removed.
- A font loaded from an external service, which blocks text from appearing until it arrives.
- A page builder plugin that ships the CSS and JavaScript for every layout it can produce, whether or not your page uses them.
Open your own homepage on a phone, on cellular data rather than your office wifi, from a cold start. If you can count past three before you can read the headline, that is the first thing to fix, and it will usually be a bigger win than a redesign.
Job two: answer the question they arrived with
People do not arrive at a local business site to browse. They arrive with a specific question, and there are usually only a handful of them: do you do this particular thing, do you cover where I am, roughly what does it cost, how soon can you come, and are you a real business.
A surprising number of sites answer none of those above the fold, and some answer none of them at all. The homepage says the company name and something about quality and dedication, the services page lists nine categories without saying what is involved in any of them, and the service area is either absent or a list of every town within a hundred miles.
On pricing specifically: you do not have to publish a price list, and for a lot of trades a price list would be actively misleading. But you can nearly always say something honest about how pricing works — that you quote after a site visit, that most jobs of a certain type land in a range, that estimates are free. Saying nothing at all reads as expensive, because the visitor assumes the worst and moves on to the next result.
Job three: make the next step obvious and thumb-sized
Once somebody is convinced, they need somewhere to go, and they need it to be reachable without pinching, zooming or hunting. In practice that means a small number of specific things:
- The phone number is a tap-to-call link, not text somebody has to copy. On a phone, a number that is not a link is a number that does not get dialled.
- There is a text option if you answer texts. For a lot of trades and services this converts better than a form, because the customer can send it in ten seconds from a job site and get back to what they were doing.
- Every tap target is at least forty-four pixels tall. Anything smaller is a coin flip with a thumb.
- The form is short. Name, one way to reach them, and what is going on. Every additional required field costs you submissions, and asking for budget up front costs you a lot of them.
- There is one dominant action per page, not five competing ones. A page with five equally weighted buttons has effectively none.
It is also worth checking where those actions actually go. Forms that send to an address nobody monitors are common enough that it is worth submitting your own form once a quarter to confirm it still arrives.
Job four: be readable by the machines, not just the people
The last job is the one that has changed most recently. Your site is read by search engines, by the system that builds your Google Business Profile knowledge panel, and increasingly by AI assistants answering questions like "who does emergency plumbing near me and are they any good".
Those systems cannot read a picture of your opening hours. They cannot infer your service area from a photo of a van. What they can read is text, page structure and structured data — a standard format that states, in a way software can parse without guessing, that you are a business of a certain type, in a certain place, offering certain services, reachable at a certain number.
This is not exotic work, and it is not a growth hack. It is closer to labeling your stock properly. But it is the difference between a system that describes your business correctly and one that guesses, and the guesses are not always flattering.
The order to do this in
If you are working through your own site, the sequence that tends to pay off is: make it fast, make it answer the obvious questions, make the next step tappable, then make it machine-readable. Design sits underneath all four — a clear, current-looking site earns trust that a dated one has to argue for — but design is the thing that makes the other four land, not a substitute for them.
If you would rather someone just looked at it and told you which of the four is costing you the most, that is a reasonable thing to ask for.