The Emergency Service Page That Actually Gets The Call

An emergency service page is read by somebody in a bad mood, on a phone, with one hand, standing somewhere unpleasant, in under sixty seconds. Every decision on that page should follow from that sentence. Most emergency pages instead read like a brochure with the word “emergency” in the heading, and they lose the call to whichever company put a tappable phone number where a thumb could reach it.

The state the reader is in

Start with an accurate picture of the person you are writing for, because it invalidates most of what normally goes on a website.

They have a problem that is actively getting worse: water where it should not be, no heat in January, a breaker that will not reset, a blocked line with guests arriving. They have already decided to pay somebody. They are not comparing your values against a competitor’s values. They are looking for two facts, in this order: can you come, and how do I reach you.

They are on a phone. They may be holding a torch. The screen may be wet. They have one hand free. They have possibly already tried two other companies and got voicemail.

Everything that belongs on an emergency page follows from that. Everything that does not serve that person is in their way.

The order of the page

There is a correct order, and it is not the order most sites use.

1. The confirmation, in one line

The first line confirms the two things they came for: what you handle and where. “Emergency plumbing across [area], answered by a person.” One sentence. No heading above it that says “Welcome”.

2. The phone number, as text, in a tel link

This is the whole page. It should be the largest interactive element, it should sit in the lower half of the first screen where a thumb reaches naturally, and it should be a real anchor using the tel scheme so that tapping it dials.

Three failure modes to avoid, all of them common:

The number as an image. It cannot be tapped, cannot be copied, cannot be read aloud by a screen reader, and cannot be parsed by a search engine. It is invisible to exactly the people and systems that most need it. If there is one thing to fix on an existing emergency page today, it is this.

The number only in the header. A sticky header number is good. A number that exists only in a 14px header strip and nowhere in the body of the page is a number that a person scrolling past will not find when they need it.

The tap target too small. A phone link set in body text inside a paragraph is a small target for a wet thumb. WCAG’s target size guidance gives a minimum for a reason, and an emergency page is the page where that minimum matters most. Give the call action a real button with generous padding.

3. The hours a person actually answers

This is where honesty pays. There are three truthful positions and all of them are fine to publish:

• A person answers live, twenty four hours, every day.
• A person answers live until a stated hour, and after that an answering service takes details and a technician calls back.
• A person answers during business hours and emergency calls outside those hours are handled on a rota with a stated call-back window.

What is not fine is the fourth position, which is what most sites do: the word “24/7” in a banner with no explanation, when in practice a call at midnight rings out. That produces a furious voicemail and a review you cannot remove.

Write the true version. Somebody deciding between you and another company at eleven at night will pick the one that told them what would happen.

4. Where you actually come to

An emergency page with no service area is a page that generates calls from outside your radius, which wastes your dispatcher’s time and the caller’s. List the towns. Link to the area pages if you have them.

If your response time genuinely differs across the area, say so. “Same hour in [town], usually within two hours in [towns]” is more useful than a single promise averaged across a county.

5. What it costs to come out

The callout fee, the diagnostic fee, the after-hours surcharge: publish them, and say what each one covers.

The argument against publishing is that it scares people off. In practice the people it scares off are the ones who would have asked on the phone and then declined, so the cost is a conversation you did not have to have. The benefit is the caller who books without hesitation because they were told up front, and who does not feel ambushed when the invoice arrives.

If you cannot publish a fixed number, publish the shape of it. “A diagnostic visit is [amount] and covers the first hour on site. If you go ahead with the repair, it comes off the total.” That is a real answer.

6. What you do not handle

A short, plain list of what is not an emergency for you saves everybody time. A plumbing company that does not do gas work should say so. An electrical contractor that does not do utility-side faults should say so, and point at who does.

This is counter-intuitive and it works, because every call you turn down is a call your dispatcher took instead of taking a job you wanted.

7. Everything else, below the fold

Credentials, licence numbers, insurance, how long you have been trading, what a technician will do when they arrive, photos of the vans. All of it is useful and none of it is why the page exists. It goes underneath, for the reader who is calmer than you assumed, or who is choosing on behalf of a property they manage rather than standing in the problem themselves.

Speed is part of the content

An emergency page is opened in a basement with one bar of signal, in a car park, in a stairwell. Page weight stops being an abstract quality metric and becomes the difference between a page that opened and a page that did not.

The general case for speed is well documented, and web.dev’s material on why speed matters covers the user behaviour behind it. For an emergency page the argument is simpler and harsher: a page that has not painted is a page with no phone number on it, and the person holding the phone will go back to the results and tap the next company.

Practical implications for this one page:

• Nothing above the fold should depend on a script finishing.
• The phone number and the hours should be in the HTML, not injected by JavaScript, so they exist even if a script fails.
• Hero imagery should be light or absent. An emergency page does not need a photograph of a smiling technician; it needs to open.
• Third-party widgets, chat bubbles and tracking that block rendering are worse here than anywhere else on the site.

Google’s mobile-first indexing guidance is a reminder that the mobile version of the page is the version that counts for search as well, which aligns the commercial and the technical argument neatly: the fast, simple, phone-shaped page is the one that both ranks and converts.

The form, and why it is not the main event

Some people will not call. Somebody in a shared office cannot take a call in a meeting. Somebody with a hearing impairment may prefer text. Somebody whose emergency is bad but not immediate might rather type it out at eleven at night than wake up a rota technician.

So the page needs a form. It just does not need to compete with the phone number.

What a good emergency form looks like:

• Short. Name, number, address or postcode, one line on what is happening, and one question about whether it is getting worse right now.
• Honest about timing. “We answer these in the morning” is better than silence and better than pretending it is the same as calling.
• Routed somewhere real. A form that lands in a website inbox nobody opens until Tuesday is worse than no form at all, because it looks like a promise.

Writing the copy

Three rules cover most of it.

Short sentences. A reader under stress does not parse long clauses. Say one thing per sentence.

Concrete nouns. “Burst pipe”, “no heat”, “tripping breaker”, “blocked main line”. These are the words the reader is thinking in, and they are also the words they typed into the search bar.

No hype. “We are passionate about delivering exceptional service” is invisible to somebody standing in a wet hallway. “Somebody picks up the phone until 10pm, and after that a technician calls you back within the hour” is the sentence that gets the job.

A structure you can copy

Here is the skeleton, in order, for an emergency page for any of the trades where emergencies exist:

1. One-line confirmation of what you handle and where.
2. Large call button with the number as text, in a tel link.
3. Answering hours, stated plainly, including what happens outside them.
4. Service area, with the towns named.
5. Callout or diagnostic fee, with what it covers.
6. Secondary form, short, with an honest response expectation.
7. What is not an emergency for you, and where to go instead.
8. What happens when a technician arrives, step by step.
9. Credentials, licensing and insurance.
10. Links to the related non-emergency service pages, for the reader who arrived here by accident.

Nine and ten matter more than they look. Plenty of emergency page traffic is not an emergency at all: it is somebody who searched urgently, landed here, and actually wants a quote for a planned job. Give them a door out of this page and into the right one.

What changes by trade

The skeleton is the same across trades. The specifics underneath it are not, and the specifics are what makes a reader believe the page was written by somebody who does this work.

Plumbing

The emergency is nearly always water going somewhere it should not, and the first thing the page can do is help. A short, calm line about where the stop tap usually is, and that turning it off before the van arrives will limit the damage, is the single most useful sentence a plumbing emergency page can carry. It costs you nothing, it demonstrably helps, and it is the kind of thing a reader remembers when they choose who to call back for the planned work later.

After that, the page should separate the three cases that need different responses: an active leak, a total loss of water, and a blocked line that is backing up. They have different urgency and different diagnostic paths, and a caller who can say which one they have gets a better answer faster.

Heating and cooling

The emergency here is seasonal and the page has to change with it. In January, no heat in a house with a young child or an elderly resident is a different call from a system that is running but noisy, and the page should say plainly which cases you prioritise. In July the equivalent is no cooling in extreme heat.

A heating and cooling emergency page should also be honest about parts. If a repair depends on a part that may not be on the van, saying so on the page prevents the conversation where a customer expects the job finished tonight and it finishes on Thursday. State what a temporary fix looks like when the permanent one has to wait.

Electrical

Electrical emergencies split cleanly into the ones a contractor handles and the ones the utility handles, and most callers cannot tell the difference. A page that explains that split in two sentences, and names who to call for the utility side, saves the caller time and saves your dispatcher a call they were going to decline.

The other thing an electrical emergency page should carry is a plain safety line: what not to touch, and when to leave the building and call the fire service rather than an electrician. Publishing that is the right thing to do regardless of whether it wins work, and it reads as competence.

Cleaning and restoration adjacent work

Not every trade has emergencies, but some have deadlines that behave like them. A turnover clean before a same-day check-in and a post-flood clean after a plumber has finished are both urgent in a way a recurring route is not. If you take that work, it deserves its own page with its own response window stated, rather than being buried in a list of services.

Measuring whether it works

Two numbers tell you nearly everything, and neither requires a complicated analytics setup.

Taps on the call link. If the phone link is a real anchor, it can be counted. If taps are low relative to page views, the number is not prominent enough or the page is not loading fast enough to be seen.

What the dispatcher says. Ask whoever answers the phone what callers mention. If callers routinely ask something the page answers, the page is not answering it clearly enough. If callers routinely ask something the page does not answer, you have found the next paragraph to write.

That second one is not a metric in any dashboard and it is the most valuable feedback loop a trade website has. The person on the phone hears the questions the website failed to answer, all day, every day.

The version most sites have, and the version they need

The typical emergency page today: a stock photograph of a technician, a heading with the word “emergency” in it, three paragraphs about commitment to quality, a list of every service the company offers, a contact form with six fields, and a phone number in the header in small text, sometimes as an image.

The version that gets the call: one line confirming you come here and handle this, a large tappable number, the hours somebody actually answers, the towns you cover, what it costs to come out, and a short form for the people who cannot call. Everything else below that line.

The second version is easier to build than the first. It just requires deciding that the page belongs to the person in the basement rather than to the marketing copy, and then removing everything that does not help them.