What a Website for an Electrician Actually Costs in 2026

TL:DR: A seven-person electrical contractor was paying £89 a month to a website builder for a five-page site they could not edit, plus £40 a month for a separate booking tool that did not talk to it. The rebuild cost a one-off fee and now runs on about £6 a month of hosting. The site is not fancy. It loads in under a second on a phone in a van with two bars of signal, which is the only performance metric that ever mattered for this business. Full breakdown of the costs, the mistakes, and the bits I got wrong below.

Every client on this site is anonymised. No names, no logos, no screenshots that identify anyone. The numbers and the technical detail are real, taken from work actually delivered. The work itself, with the numbers, is on our outcomes page, described by sector rather than by name.

This one is a seven-person electrical contractor in the English Midlands. Domestic and light commercial, mostly rewires, consumer unit upgrades, EV charger installs.

What they had before

Three separate things that did not know about each other.

That is £1,716 a year. For a five-page site.

The worst part was not the money. It was that the booking tool sent enquiries to an address the office manager did not have on her phone, so jobs sat unread over weekends. They found out because a customer rang up annoyed that nobody had got back to them for four days. That is the actual cost, and it does not show up on any invoice.

What we built

Nothing clever. That is the point.

• A static site, six pages, generated at build time. No database on the public side.
• A form that posts to a small endpoint, writes the enquiry to a database, and sends a copy to two real inboxes and a phone notification.
• A page per service area, because "electrician in [town]" is what people actually type.
• Photographs of their own vans and their own work, taken on a phone, not stock images of models in clean overalls holding a clipboard.

The whole thing is hosted on a small VPS. Hosting works out at roughly £6 a month, and that box also runs two other things, so it is really less than that. If you want the maths on cheap hosting, I wrote it up in the Hetzner and Coolify guide.

The bit that mattered most

Load time on a bad connection.

Their customers are on driveways and building sites. Half the traffic came from phones on mobile data, often poor mobile data. Ofcom's Connected Nations report is a useful reality check: outdoor 4G coverage from all operators is still not universal across the UK, and "has signal" is not the same as "has usable bandwidth". The old site was pulling nearly four megabytes of page builder scripts, a chat widget nobody answered, and three tracking pixels.

We got the homepage under 300KB. On a throttled connection that is the difference between a page appearing and a customer pressing back. Google publishes the thresholds it cares about in the Core Web Vitals documentation, and they are worth reading before anyone sells you a redesign on the basis of how it looks on their desktop monitor.

What it cost

One-off build fee, then about £6 a month for hosting and £1 a month for the domain, plus email.

I am not going to publish the build fee, because it varies enormously with scope and publishing one number invites people to treat it as a price list. What I will say is that it was less than a year of what they were paying before, and there is no monthly fee to me afterwards. They own the code. If they want to take it elsewhere, they can, and I will hand over the repository without an argument.

Email moved to a proper provider rather than the bundled one. That fixed the spam flagging, which was a deliverability problem, not a website problem. I compared the options in the email hosting write-up, and the current prices for all of them are in the pricing index.

What broke

Two things, and I would rather write them down than pretend the job went perfectly.

The form silently failed for nine days. The endpoint was fine. The notification was going to an email address with a typo in it, so enquiries were landing in the database and nowhere else. Nobody noticed because the database was filling up correctly, so every dashboard I looked at said the system was healthy. I only found it when I checked the actual inbox rather than the actual database. Embarrassing, and entirely my fault for testing the thing I built rather than the thing the customer touches.

Now there is a check that sends a test enquiry through the real path once a day and shouts if it does not arrive in the real inbox.

The service-area pages were too thin at first. I generated one page per town with the same three paragraphs and the town name swapped in. That is exactly the doorway-page pattern Google's spam policies tell you not to do, and it deserved to not rank. We cut them down to the areas they genuinely cover and wrote each one properly, with real jobs done in that area. Fewer pages, all of them real.

Would I do it the same way again

Mostly.

The static site was right. The single VPS was right. Owning the code was right.

What I would change is starting with the enquiry path and working backwards. I built the site first and bolted the form on near the end, which is how the typo survived nine days. For a trades business the form is the product. The rest of the site is just what convinces someone to fill it in.

I would also stop assuming anyone will log into anything. The office manager was never going to open a dashboard. She lives in her email and her phone. Any system that needed her to visit a website to see a new job was going to fail, no matter how nice the website was. This is the same reason most small businesses quietly abandon a CRM within a month, which I have gone on about at length in why most project management tools suck.

Is this worth it for your business

Honest answer: only if you are already paying a monthly fee for something you cannot edit.

If you have a free Google Business Profile, a phone number that gets answered, and you get all your work through word of mouth, a new website is not your bottleneck and anyone telling you otherwise is selling something. For a domestic electrician, being on a competent person scheme register such as NICEIC sends more qualified enquiries than most redesigns will. If it is the monthly software bill bothering you, start with cutting the subscriptions instead. Google's own guidance on Business Profiles will do more for a local trade than most redesigns.

If you are paying £100 or more a month for a site you have to ring someone to change, that is worth fixing, and it usually pays for itself inside a year.

You can see the kind of thing I build on the features page, and the free self-hosted tooling is on compare.

FAQ

How much does a website for an electrician cost in 2026?

It ranges from nothing on a free Google Business Profile to several thousand for a custom build. The monthly cost matters more than the build cost. This client went from £143 a month to about £7 a month, so the rebuild paid for itself well inside a year, which is the only calculation that really matters.

Do I need a booking system or is a contact form enough?

For most trades under about ten people, a form is enough, provided the enquiry actually reaches a human within minutes. A booking system that writes to an inbox nobody watches is worse than a form, because it feels handled when it is not.

Should I build service-area pages for every town nearby?

Only for areas you genuinely cover and can write about specifically. Generating dozens of near-identical town pages is a doorway-page pattern that Google's spam policies explicitly target. Fewer, real pages outperform many templated ones.

Why does site speed matter for a trades business?

Because your customers are on phones, often on poor mobile connections, standing outside. A four-megabyte page built for a desktop preview simply does not load for them, and they press back. Speed here is not a vanity metric, it is whether the page appears at all.

Do I own the website if you build it?

On my builds, yes. The code goes in a repository the client controls and there is no monthly fee to me afterwards. If you are quoted a monthly fee that includes "the website", ask specifically what happens to the site if you stop paying.

What is the most common mistake you see on trades websites?

The enquiry going somewhere nobody checks. I have done it myself, as above. Test the path your customer actually takes, not the system you built.

Bottom line

A seven-person electrical contractor was paying £1,716 a year for a five-page site they could not edit and a booking tool that lost enquiries over weekends. The rebuild put them on roughly £84 a year of running costs, fixed the email deliverability, and made the site load on bad mobile data.

It is not a glamorous project and there is no dramatic revenue chart at the end of it, because I did not measure their revenue and I am not going to invent a number. What I can tell you is what it cost, what broke, and what I would do differently.

If you are a trade paying a monthly fee for a website you cannot change, that is the thing worth looking at first.