Why every other rule inverts
Most marketing advice assumes a considered purchase: somebody researches, compares, builds trust over weeks, and eventually makes contact. Nearly everything written about trade websites assumes that, including the rest of this cluster.
The emergency customer does none of it. There is water coming through a ceiling, or no heating with a baby in the house, or a tile off and rain forecast. They search once, they look at the top two or three results, and they ring. Total elapsed time: under a minute.
What that does to every normal recommendation:
- Trust-building content is irrelevant. They will not read it. They are judging you on whether you answer.
- The contact form is worse than useless. Nobody fills in a form at eleven at night with the mains off, and offering one as the primary action reads as a business that is closed.
- Brand design does not participate. They are looking at a map listing, not your homepage.
- The homepage may never be seen at all. They will land on whatever page matched, or on the profile, and tap the number from there.
- Speed stops being a ranking consideration and becomes the whole thing. Four seconds is a lost job.
Which means the emergency route has to be designed on its own terms rather than as a line on a services page.
It is won in the map pack
The local results block sits physically above the organic results. An emergency customer does not scroll past it, so for this kind of work the Google Business Profile is not a supporting channel — it is the channel.
What matters on it, specifically for emergencies:
- Hours that are true, including out-of-hours. If you genuinely answer at 2am, say so and set the hours to reflect it. If you do not, do not claim it.
- Service areas set to the places you will actually travel to at night, which may be a smaller radius than your daytime coverage. Being listed somewhere you will not go produces calls you turn down, and turned-down calls become poor reviews.
- The right primary category, and the emergency service listed in the services.
- Recent photographs. A profile whose last image is from two years ago reads as a business that may not still be trading, which is exactly the doubt an anxious customer resolves by ringing somebody else.
- Reviews, and particularly recent ones. "Came out at 11pm on a Sunday" in a review is worth more than anything you could write about yourself.
There is a separate guide on this site covering the profile properly, and another on getting reviews. For emergency work they are the two highest-value things on the list, and both are free.
Do not claim 24 hours unless you mean it
This is the one that does real damage, and it is almost universal.
"24 hour emergency call-out" appears on a large share of trade websites. On many of them nobody answers after nine, and the calls go to voicemail.
Consider what that costs. A customer in genuine trouble rings, gets voicemail, and rings somebody else — so you lost the job anyway. But you also used up their goodwill, and a proportion of them leave a review saying the 24-hour line is not answered. That review is then the thing the next emergency customer reads.
Publishing real hours wins more work than claiming false ones. "We answer until 10pm, seven days" is a promise you keep, and it still captures the evening and weekend work that is most of the emergency market anyway. The genuine 3am jobs are a small fraction of it and they mostly go to firms set up for them.
If you want the out-of-hours work, the answer is not a claim on a website — it is a rota, or a call-answering arrangement, or an honest statement of when you pick up. The website can only report what is true.
What an emergency page should contain
Short, and arranged in the order the customer needs it.
- The phone number first, large, as a tappable link, before anything else on the page.
- When you answer, in plain words at the top. "Answered until 10pm, seven days."
- What it costs to come out, with conditions. The customer ringing round is screening on exactly this, and a page that will not say forces a call half of them will not make.
- Where you will travel to at this hour.
- How fast, realistically. "Usually within two hours" if that is true. Vagueness here reads as evasion.
- What to do right now while they wait — where the stopcock usually is, how to isolate the supply, what not to touch. This is the one piece of genuine content that belongs on an emergency page, it costs you nothing, and it is the thing that makes a frightened customer decide you are the competent one.
What should not be on it: a form above the number, a carousel, a cookie banner covering the number on a small screen, a chat widget that takes three seconds to load, or anything that moves while the page settles.
That last one is worth repeating because it is invisible. If the layout shifts as the page finishes loading, the call button is not where the thumb expected it. The tap lands on something else, and the customer rings the next result. The Pro Pave site on this site measures a cumulative layout shift of zero, taken on the live homepage on 4 September 2026 — that is the number to aim at, and it is a build decision rather than a design one.
Speed, in the only conditions that count
Test the emergency route the way it will actually be used: on a phone, on mobile data rather than wi-fi, indoors, with one or two bars.
A site that is quick on the office connection and sluggish on a hallway signal has been tested in the wrong conditions, and the gap between those two is where emergency work is lost.
The reference figures we publish are from the Pro Pave homepage: 51 milliseconds to first byte, 0.56 seconds to first contentful paint, 0.68 seconds to full load, from eleven requests, with zero layout shift — measured on the live site on 4 September 2026 using the browser's own timing APIs. Pro Pave is a paving company, which is planned work rather than emergency work; the figures are cited for what a lean build achieves, not as emergency-trade experience.
Eleven requests is the useful part. A page-builder assembly commonly pulls over a hundred — a slider library, several font files, an animation framework, a chat widget, a cookie script, a dozen oversized images. On a strong connection you will not notice. In a hallway at eleven at night, each one is a round trip and the customer is already dialling somebody else.
The emergency job is worth more than the emergency job
Here is where the economics actually work, and it is the reason emergency work is worth chasing even at awkward hours.
A customer you rescued at midnight is not a one-off transaction. They are the person who rings you in March about the bathroom, who gives your number to their neighbour, and who writes the review that wins you the next midnight call. The emergency was the introduction.
Almost no trade captures that deliberately. The job gets done, the invoice gets paid, and nothing happens. Two things close the loop:
- Ask for the review while you are still there, or the same day. Not three weeks later by email. They are never going to be more grateful than they are right now, and a review that mentions the hour you turned out is the single most persuasive line on your profile.
- Leave the planned-work route open. A card, a follow-up a fortnight later, a sentence on the invoice. The emergency customer has just learned they can trust you, which is the expensive part of winning the planned job and you have already paid it.
And make sure the tap that produced that midnight call was counted, or you will never know which page or which profile produced it. Trade enquiries arrive by phone; if a tap on the number does not fire a tracked event, your data describes only the people who filled in a form, and the emergency customer never does.
What to check tonight
- Search your own trade plus your town on a phone, on mobile data, at nine in the evening. Where do you appear?
- Are your Google Business Profile hours true, including out-of-hours?
- Do your night-time service areas match where you will actually travel?
- Is there an emergency page, and does it lead with a tappable number rather than a form?
- Does it say when you answer, what the call-out costs, and how fast you can get there?
- Does it tell the customer what to do while they wait?
- Does a cookie banner or chat widget cover the number on a small screen?
- Does the layout move as the page loads?
- Does a tap on the number fire a tracked event?
- When did you last ask an emergency customer for a review, on the day?
The first one is the whole audit in a single action, and almost no trade has ever done it on the connection their customers are actually using.
