In short
A landing page converts when it continues the promise that brought someone there, states the offer plainly, removes doubt in the order doubts arise, and asks for one action. The strongest lever is message match — the page echoing the ad, email, or search that produced the click. The most common failure is a page serving several audiences and several goals at once.
One page, one job
A landing page differs from a homepage in a single respect: it serves one audience arriving from one place with one intent.
That constraint is the whole advantage. You know roughly what they clicked, so you can continue that thought rather than starting over. A page trying to serve three audiences forfeits it and becomes a worse homepage.
If you need to reach three audiences, that is three pages. They are cheap.
Message match decides more than design
The highest-leverage factor is whether the page continues the promise that produced the click.
Someone clicks an ad about website speed audits and lands on a general services page. The offer is technically present, three scrolls down, among nine others. They leave — not because the page is bad, but because it did not confirm they were in the right place.
The headline should echo the source. Same words where possible. That single alignment moves conversion more than most visual changes, and costs nothing.
The corollary: if you run several campaigns to one page, you have a message-match problem by construction.
The order that works
Not a template — an order of questions, answered as they arise.
1. Am I in the right place? Headline and subheading, echoing the source. Seconds.
2. What exactly is this? The offer in concrete terms. Not benefits yet — what the thing is.
3. What do I get? Outcomes, specific. "A prioritised list of what is slowing your site and what to fix first" rather than "actionable insights".
4. Why should I believe you? Proof, positioned where doubt appears. Results with numbers, a recognisable client, a relevant case.
5. What does it cost me? Money, time, or commitment. Silence here reads as expensive.
6. What happens next? Exactly what follows the click. Ambiguity at the button is friction at the worst moment.
7. What if I am wrong? The reversal — no card, cancel anytime, no obligation. Removing risk moves more people than adding urgency.
When a page underperforms, one of these is usually missing or in the wrong place.
One call to action, repeated
One action, offered more than once.
Repeat the same CTA as the page progresses — after the offer, after proof, at the end. People decide at different points, and making them scroll back loses some of them.
What does not work is several different actions. "Book a call, or download the guide, or join the newsletter" converts worse than any one of them alone, because a decision between options is harder than a decision to act.
Make the button say what happens. Book a 20-minute call beats Submit, which beats Learn more.
Proof, placed deliberately
Proof works where the doubt is, not in a band of logos at the bottom.
After a claim about results, put the result. Beside the price, the reassurance about value. Near the form, the note about what you will not do with their details.
Specific beats glowing. "Cut load time from 4.2s to 900ms; organic traffic up 34% in three months" is worth more than a paragraph of praise, because it can be checked and it describes something recognisable.
Anonymous testimonials read as invented. A name, a role, and a company are the minimum for a quote to carry weight.
Forms
Every field costs conversions. Ask only for what you need to take the next step, not what you would eventually like to know.
A booking needs a name, an email, and a slot. It does not need a company size, a budget range, and a phone number — those can be asked in the conversation you are trying to start.
If sales genuinely needs qualifying information, ask after the commitment rather than before it. A form completed in two steps, where the second arrives post-booking, converts better than one long one.
What to cut
Navigation. A landing page with a full site menu offers exits. Keep a logo linking home and little else.
Anything not serving this offer. Other services, general company history, unrelated posts.
Carousels. People rarely advance past the first slide, and everything after it is effectively unpublished.
Fake urgency. Countdown timers on evergreen offers are transparent and damage the trust you need at the decision point.
Long intros. Nobody arrives wanting context. Answer the question, then explain.
Measuring it
One primary metric: the completion rate of the one action. Everything else is diagnostic.
Useful diagnostics: scroll depth showing where people stop, which CTA position gets clicked, and drop-off within the form. Those tell you which of the seven questions failed.
Be careful with tests. Small traffic and short windows produce confident-looking results that are noise, and acting on noise is worse than not testing at all.
If you are running traffic to a page that is not converting and want a read on which question it fails, book a call.
Common questions
What makes a landing page convert?
Continuing the promise that produced the click, stating the offer concretely, answering doubts in the order they arise, and asking for one action. Message match — the headline echoing the ad, email, or search that brought someone there — is the single strongest lever and costs nothing to fix.
How many calls to action should a landing page have?
One action, repeated at several points as the page progresses, since people decide at different moments. Offering several different actions converts worse than any one alone, because choosing between options is harder than deciding to act.
Should a landing page have navigation?
No, or almost none. A full site menu offers exits from a page whose entire purpose is one action. Keep a logo linking home and remove the rest, along with anything that does not serve this specific offer.
How many form fields should a landing page have?
Only what is needed to take the next step, not everything you would eventually like to know. A booking needs a name, an email and a slot; company size and budget can be asked in the conversation you are trying to start. If qualifying information is genuinely required, ask for it after the commitment rather than before.
Where should testimonials go on a landing page?
Wherever the doubt occurs, rather than in a band at the bottom. Put a result immediately after the claim it supports, reassurance about value beside the price, and privacy wording near the form. Specific, attributed results outperform anonymous praise, which reads as invented.
