The decision takes under a minute
Here is the actual sequence, because most plumbing websites are designed as though it were something else.
A valve fails at seven in the evening. The customer turns the water off, gets their phone out, searches, and taps the first two or three numbers in the map pack. Whoever picks up gets the job. The entire decision takes less time than it took to read this paragraph.
Nothing on a conventional website participates in that. Not the hero image, not the About page, not the testimonial carousel, not the contact form. The only things that matter are: are you visible, is the number one tap away, and does somebody answer.
So the first job of a plumbing website is extremely narrow — be fast, be visible, put the number where the thumb already is. Everything else on the site is for a different customer, and it is worth being clear in your own head which pages serve which.
That second customer is the planned job: a new bathroom, a boiler replacement, heating across a whole house. Four figures, several weeks, two or three quotes compared, and a partner involved in the decision. They read everything. They want to know which boilers you fit and why, what the warranty is and who honours it, how long you will be in the house, whether you tidy up, and what happens if something fails in two years.
One site, two jobs. Most plumbing sites do a mediocre version of the second and none of the first.
The three things almost every plumbing site gets wrong
Not design faults. Structural ones, and all three are cheap to fix.
The number is not tappable
It is text inside an image, or buried in the footer, or it is a tel: link that only appears after a menu tap. At seven in the evening with the water off, that is the whole game, and a lot of sites lose it before the customer has read a word.
One page is doing nine jobs across a dozen towns
A homepage listing "boilers, bathrooms, leaks, power flushing, radiators, underfloor" across "Bolton, Bury, Rochdale and surrounding areas" competes with itself for every combination of those. Nothing ranks, because nothing on the site is about anything in particular.
Nobody is counting the calls
If a phone tap does not fire a tracked event, every figure you are shown describes the minority of customers who used a form. You cannot tell which pages produce work, so you cannot tell what to build next. It is an hour of setup and it changes every decision after it.
Publish the call-out charge
This is the recommendation plumbers push back on hardest, and it is the one with the clearest payoff.
A customer ringing round at seven in the evening is screening on two things: can you come, and what will it cost to find out. A site that answers neither forces them to ring and ask, which costs them a call and costs you an interruption, and half of them will not bother.
Publishing the call-out charge, the hourly rate or the band, with its conditions stated, does three things. It removes the awkward conversation from the start of the call. It filters out the people who were never going to pay it, before they reach your phone. And it reads as confidence — the firm that will not say is the firm the customer assumes is expensive.
The conditions are what makes it safe to publish: "£X call-out including the first thirty minutes, weekdays 8am to 6pm; £Y evenings, weekends and bank holidays; parts on top at cost plus Z per cent; no charge if we cannot get access". That is not a quote and nobody will hold you to a number you did not give. It is enough information for somebody to decide to ring you rather than the next result.
The same applies to hours. "24 hour emergency" on a site where nobody answers after nine is worse than publishing nine. The second one keeps the customer who would have been let down and would have left a review about it.
One page per job, one page per town
The structural fault is the same across every trade and it is worth stating in plumbing terms.
A homepage that lists "emergency plumbing, bathroom installation, boiler repair, boiler replacement, central heating, power flushing, leak detection, drainage" across "Manchester, Salford, Stockport, Bury and Bolton" is a single page competing against itself for forty different searches. It will rank well for none of them, because it says almost nothing about any one of them.
The alternative:
- A page per job type, each answering what it involves, how long it takes, what it typically costs and what can go wrong. "Boiler replacement" deserves a page with real content on it, not a line in a list.
- A page per town you genuinely work in, naming real jobs done there. Not a footer list of thirty place names, which is the pattern Google has been discounting for years and which reads as padding to a human too.
- A page per question a customer asks before they ring — what a power flush actually does, how long a boiler should last, whether a leak is covered by insurance, what a landlord gas safety check involves. These rank easily and they build the trust that makes somebody pick you for the planned job later.
- Evidence: recent jobs, dated, photographed, in named places.
Pro Pave, a paving company on this site working across Greater Manchester, needed twenty-seven pages of services and locations on exactly this reasoning. Different trade, identical problem: several distinct services, a spread of towns, and a homepage that cannot carry any of it properly.
Twenty-seven is not a target. One page per search somebody actually makes is the rule; the number falls out of your own service list and coverage.
Measured, not described
Site speed is abstract until you put it next to the emergency customer standing in a kitchen with the mains off.
Four seconds is the difference between your number and the next one. And the test that matters is not the one run on an office connection — it is mobile data, often indoors, often poor.
Rather than describe what good looks like, here are our own figures. The Pro Pave homepage answers in 51 milliseconds, shows content in 0.56 seconds and loads completely in 0.68 seconds, from eleven requests, with a cumulative layout shift of zero. Measured on the live site on 4 September 2026 with the browser's own timing APIs.
The eleven requests is the design decision underneath the number. A bespoke theme loads what the page uses. A page-builder assembly loads a slider library, three font families, an animation framework and a dozen desktop-resolution images whether the page needs them or not, and each is a round trip on a bad connection.
The zero layout shift is the one that quietly costs plumbers money. If the page is still settling when the thumb comes down, the call button has moved and the tap lands somewhere else. The customer does not try again. They go back and ring the next result, and nothing in any report you are shown will ever tell you it happened.
If you are not counting taps, you are flying blind
Plumbing is the clearest case of this in any trade. Your enquiries arrive by phone because the alternative is typing a form with wet hands.
If a tap on your phone number does not fire a tracked event, then every number in every report describes only the minority of customers who used a form — typically the planned-work customer, almost never the emergency one. You will conclude your bathroom pages work and your emergency pages do not, and you will be exactly wrong.
It is a few lines of code on the phone link, it takes an hour, and almost no plumbing site has it. Once it is running, the question "which page produced that call" becomes answerable, and the answer usually redirects the next six months of work.
What to check on your own site
- Is the phone number a tappable link in the header, above the fold, on every page, on mobile?
- Does tapping it fire a tracked event?
- Is the call-out charge or rate published, with its conditions?
- Are your real hours published, rather than "24 hour" on a line nobody answers at midnight?
- Does each job type have its own page with actual content, or is it a list item?
- Are your towns pages with real jobs on them, or names in a footer?
- Is there depth for the planned-work customer — boilers, bathrooms, heating — separate from the emergency route?
- How fast is the homepage on mobile data, measured?
- Does the layout move while it loads?
- Is your Google Business Profile complete, with true out-of-hours and recent job photographs?
Numbers one, two and three are a morning's work between them and they are where most plumbing businesses are losing calls right now.
