Guide

What a buyer decides about your website before they scroll

Somebody lands on your site from a search on their phone, at the point in the day when they have decided to deal with the thing. They are not reading. They are running four questions in a few seconds, in a fixed order, and if any of them fails they go back to the results page and try the next one. None of the four are about how good you are at your job. This is the written companion to our website test, which scores a page against exactly these judgements.

The four questions, in order

The order matters more than the list, because a failure early makes everything after it irrelevant. A beautiful proof section does nothing for a visitor who already decided you do not do their kind of work.

  1. Is this the right kind of business? Answered by the first heading, in about a second. Not your name, what you do, and for local work, where. This is the single most common failure and it is the most expensive.
  2. Is it real and still trading? Answered by peripheral vision, not by reading: a phone number, an address, hours, recent-looking dates. A stale copyright year reads as a closed business faster than any amount of good copy repairs.
  3. Can I do the thing I came to do? One obvious action. Call, book, or ask. If the only route is a contact page two clicks away, a meaningful share of people never take it.
  4. Is there any reason to pick this one? Only reached if the first three passed. This is where reviews, named clients and specifics earn their place, and it is the only question most websites are actually written to answer.

Why the heading does most of the work

If you change one thing after reading this, change the first line of your homepage.

The most common headline on a local business website is the business name, sometimes with a tagline underneath. To the owner it is obvious what the business does. To a stranger who arrived from a search, it is a name they have never heard, attached to nothing.

  • Name the trade in the reader's words. What they typed, not what your industry calls itself.
  • Name the place, if you serve one. A local visitor is checking whether you cover them before anything else.
  • Add the qualifier that filters. Same day, by appointment, commercial only, no callout charge. Filtering out the wrong customers is a feature, not a loss.
  • Keep it a sentence a person would say. If you would not say it to someone at a counter, it is not a heading, it is a slogan.
The same business, two first lines
What the heading saysWhat the visitor learnsWhat happens next
Beaulieu and SonsA name, and nothing elseBack to the search results
Quality service since 1994This could be any business in any tradeBack to the search results
Emergency plumbers in Laval, same dayThe trade, the place, the urgency, all at onceThey look for the phone number

The phone number is a bigger deal than it looks

For a local business, the phone number is not contact information. It is the conversion button, and it fails in three specific ways that are all invisible to the owner, because the owner already knows the number.

It is in an image. Common on older sites and on anything designed as a graphic. It cannot be tapped, cannot be copied, cannot be read by a search engine, and cannot be enlarged by someone who needs to.

It is text but not a link. On a phone, that means the visitor has to memorise it, leave, open the dialler and type it. Every one of those steps loses people, and the fix is a single attribute.

It only appears in the footer. Which means it exists for people who already decided, and is invisible to everyone still deciding.

What actually counts as proof

Question four is where most of the effort goes, and most of that effort goes into the wrong things. Proof is anything a stranger cannot dismiss as something you wrote about yourself.

  • Counts: reviews with names attached, named clients, photographs of your own work, a real address, a person's face, specifics about jobs you have done.
  • Does not count: "trusted by hundreds", stock photography of an office nobody works in, badges nobody recognises, adjectives about your own quality.
  • Counts more than people expect: saying who you are not for. A business that turns work down reads as a business with enough work, and it is the fastest credibility you can buy with a sentence.
  • The order matters here too: two specific reviews beat twenty generic ones, because the specific ones cannot have been written by you.

The nine checks, and why those nine

Our website test scores a page out of 100 across nine fixed rules. They are deliberately mechanical: a rule gives the same page the same score every time, and there is nothing for a model to invent. They map onto the four questions above.

The nine checks and the question each one is really asking
The checkThe buyer question behind it
The first screen says what you doIs this the right kind of business
The phone number is text, not a pictureIs this real, and can I act on it
The phone number is tappable on a phoneCan I do the thing I came to do
There is an obvious way to book or askCan I do the thing I came to do
Getting in touch takes no huntingCan I do the thing I came to do
A real business is visible behind the siteIs it real and still trading
Nothing on the page looks abandonedIs it real and still trading
Someone other than you vouches for youIs there a reason to pick this one
The page is not heavy before it startsDo I get to judge any of this before I give up

How to run this on your own site properly

Not by looking at it. You cannot see your own site cold, because you know what it says.

  1. Score it mechanically first. Run the website test and fix everything it fails. That is the floor, and it is all fixable in an afternoon.
  2. Then get a real four seconds. Show the homepage to someone who does not know the business, on a phone, for four seconds. Take it away and ask them what the business does and where. Their answer is your heading's real score.
  3. Do it on a phone, on data, not on your desk. Most local traffic is mobile and often on a poor connection. A page that is fine on a laptop can fail question one entirely on a phone because the heading sits below a full-screen image.
  4. Fix in the order of the four questions. Heading first, then the evidence you are real, then the action, then the proof. Fixing question four while question one fails changes nothing.

FAQ

Is four seconds a real measurement or a figure of speech?
It is a working figure, not a citation, and we use it because it matches what we see when we put people in front of unfamiliar sites. Treat the number as shorthand for "far less time than you think" rather than as a research result. The order of the four questions is the part that matters and the part that holds up.
Our site scored well and we still get no enquiries.
Then the floor is not your problem, which is genuinely useful to know. The next places to look are whether the right people are arriving at all, and what happens after they get in touch. Both are more work to fix than a heading, which is why it is worth eliminating the cheap causes first.
We are a referral business. Does any of this matter?
More than you would think. A referred customer still looks you up before calling, and they arrive with a name and no context, which is exactly the visitor question one is written for. A referral that lands on a page that does not confirm what you do gets quietly dropped.
Should we redesign?
Almost never, based on this. Every one of the nine checks is a small change to an existing page: a heading, a link, a footer, a stale year. Redesigns are how a two hour fix becomes a three month project that never ships.
Does this apply to ecommerce?
The four questions do, in a different order: shoppers check whether you have the thing before they check whether you are real. The phone-specific checks matter less, and delivery, returns and stock become the trust signals. Most of the rest transfers directly.

Try the tool