Most landing pages are built to describe, then judged on conversion. Those are different jobs. A description can be complete and still leave a visitor with three unanswered questions, and unanswered questions are why people leave.
Conversion work is subtraction. For every element, ask what doubt it removes. If it removes none, it is decoration competing with the button.
Match the promise that brought them
If an ad offered same-day boiler repair, the page heading should say same-day boiler repair. Not heating solutions for modern homes. A visitor who has to check whether they are in the right place will often leave rather than check.
Repeat the specific words from the source: the service, the location, the offer, the deadline. This is the cheapest improvement available and it is routinely skipped because the page was written before the campaign.
Answer five hesitations
Most buyers hesitate for one of five reasons. Is this for someone like me? What does it cost? Can I trust you? What happens after I click? How much effort will this take?
Answer each in plain language, near the decision rather than in a footer. Who it is for, with an example. A price, or the reason a price is not shown and what happens instead. Proof: a review, a photograph of real work, a name and a face. The next step, described in one sentence. And the effort involved, including how long it takes and what the visitor needs to prepare.
Visitors do not need to be persuaded. They need to be reassured that the thing they already want will not waste their time.
Structure the top of the page
Above the fold, four elements: a headline stating the outcome, a subheading naming who it is for, one piece of proof, and the action. No carousel. No stock photograph of a handshake. No paragraph explaining the company's history.
The headline should be readable at a glance on a phone. If it wraps to four lines, it is too long. Put the detail underneath, where people who want it will scroll.
Reduce the form to what you need
Every field costs completions. Ask for a name and one contact method, plus the one question that lets you reply usefully. Collect the rest on the call.
Label fields above the box rather than inside it, so the label does not vanish when typing starts. Show errors next to the field, in words a person would use. Confirm what happens after submission on the button and in the message that follows.
Test the form yourself on a phone. Submit it once. Check where the notification lands and how quickly. A form that works but sends enquiries to an unread inbox is not working.
Speed is part of the argument
A slow page tells the visitor that the business is slow. Compress images, remove scripts that do not earn their place, and check the load on a mid-range phone over mobile data rather than office wifi.
The most valuable element on a mobile page is often the phone number. Make it tappable and keep it visible without scrolling, especially for services where people want to speak to someone before deciding.
Change one thing at a time
A landing page with five changes and a better result teaches you nothing, because you cannot tell which change worked. Pick the element most likely to be causing hesitation, change it, and run long enough to see a pattern.
If traffic is low, testing is unreliable. In that case, use five conversations instead: ask five people who did not convert what stopped them. The answers are usually more specific than any split test.
The conversion checklist
- The headline repeats the promise from the ad or post.
- One action per page, repeated at the top and the end.
- Who it is for stated with a concrete example.
- Price, or an explanation of what happens instead.
- Proof above the fold: review, photograph or named client.
- Form reduced to the minimum needed to reply.
- Page tested on a phone, including a real submission.
- Load time checked on mobile data.
- One change per test, with a note of the date and result.
The strongest landing pages read as if written for a single customer with a specific problem. That narrowness feels risky. It is the opposite: a page for everyone gives no one a reason to act, while a page for one person gives that person a reason to click.
