Landing pages built to convert

One goal per page, and what that rules out

A landing page converts because it removes choices. The visitor arrived with one question in mind, and the page answers that question and asks for one thing. Every additional option splits attention and lowers the chance of any single action happening.

One goal does not mean one button. It means one outcome. A page can offer the same conversion in two forms, a form and a phone number for example, and still be single-goal. What breaks it is offering genuinely different outcomes on the same page: book a call, download a guide, subscribe to the newsletter, read the blog, follow on social. Each of those pulls a different type of visitor in a different direction, and the page ends up serving nobody well.

The discipline shows up in what you leave out. If a section does not move someone toward the single action, it is costing you scroll depth. That includes company history, a full service list, and awards that have nothing to do with the offer. The homepage is where a business explains everything. A landing page exists for exactly one promise.

• Decide the single outcome before writing a word of copy
• Repeat the same call to action down the page rather than adding new ones
• Cut sections that do not push toward that outcome
• One offer per audience, so build separate pages instead of one hybrid
• If you cannot name the goal in one sentence, the page is not ready

Message match: continuing the promise that earned the click

Message match means the page picks up exactly where the ad, email or social post left off. Same promise, same words, ideally the same visual. When it breaks, the visitor experiences a moment of doubt about whether they landed in the right place, and a large share of them leave in that moment rather than reading on.

The most common break is scope. An ad promises something specific and the page talks about the whole company. Someone who clicked an ad for emergency plumbing in Leeds should land on a page whose headline says emergency plumbing in Leeds, not a page headed Welcome to our family business. The second common break is offer drift, where the ad names a price or an incentive and the page never mentions it. That reads as a bait and switch even when it was just an oversight after the offer changed.

This is also why campaign pages need maintenance. Ads get edited, offers expire, seasonal promotions end. A quarterly pass comparing live ad copy against live page copy catches drift before it quietly costs you a quarter of your paid budget.

• Echo the exact phrase from the ad in the page headline
• Carry the same image or creative style across from the ad
• Name the same offer, discount or guarantee the ad promised
• Match the specificity: a location-targeted ad needs a location-specific page
• Re-check page copy every time the ad copy changes

What has to be above the fold, and what does not

Above the fold matters less than people think and more than the contrarians claim. It does not need to contain everything. It does need to answer three questions fast: what is this, is it for me, and what happens next. If those three land, the visitor keeps scrolling, and the rest of the page gets its chance.

A clear headline stating the offer, a subheading that adds the qualifier, and a visible primary action usually cover it. The visual matters here too. A photograph of a real person, product or workplace outperforms a generic stock image, because a stock image reads as a template and quietly signals that the rest may be generic too.

What does not need to be above the fold is the form itself, unless the offer is very low commitment. For a considered purchase, asking for details before making the case is premature. People do scroll when the opening earns it, and a form placed after the proof often outperforms the same form pushed to the top. The test that matters is whether the opening screen makes the case for reading on. On mobile that opening screen is small, so design it there first and let the desktop version inherit the discipline.

• Answer what it is, who it is for, and what to do next, in the first screen
• Use a real image of the product, team or work, not generic stock
• Design the mobile first screen first, it is the tightest constraint
• Place the form after the proof for considered purchases
• Avoid opening on a large slider, it delays the message and slows the page

Form friction: every field has a cost

Each field on a form is a small tax on completion. Some of those taxes are worth paying because the information genuinely qualifies the lead. Many are collected out of habit, and those are pure loss.

Go through the form field by field and ask who uses this data and what decision it changes. Phone number when nobody calls. Company size when the offer is the same either way. A dropdown of how did you hear about us that nobody reads. Fields like these get cut, or moved to a later stage after the person is already committed. Where qualification genuinely matters, more fields can be correct, because fewer but better leads beats a pile of enquiries the sales team cannot use. The right number depends on what you do with the answers, not on a universal rule.

Beyond field count, the mechanics matter. Set correct input types so mobile keyboards switch to numbers or email. Validate as people type rather than only on submit. Keep every entry when validation fails. Say what happens after submission, because uncertainty about being called at a bad moment stops people more often than the form length does. And check the page after submission, because a success message that just says thank you wastes the moment when attention is highest.

• Cut any field whose answer changes nothing you do
• Set input types so mobile keyboards match the field
• Validate inline and never wipe what the user already typed
• Say what happens next: who contacts them, how, and how soon
• Use the thank-you page for the next step, not just a confirmation

Social proof that is honest and specific

Proof works when it is specific and verifiable. It stops working when it looks assembled. A row of unbranded logos, a testimonial signed with initials, a star rating with no source: these read as decoration, and an experienced buyer discounts the whole page because of them.

Strong proof names the person, the company and the situation, and describes something concrete that happened. Weak proof praises in the abstract. A quote saying the team was great is worth less than a quote explaining what problem was solved and what changed, because the second one is hard to fabricate and easy to picture. Where you have review platform ratings, link to the source so the claim can be checked. Where you have results, describe what was measured and over what period rather than presenting a bare number with no context.

If you do not yet have proof, do not manufacture it. Placeholder testimonials and invented results are a serious credibility risk and, for review content in particular, a regulatory one in several markets. Use what you honestly have instead: a clear guarantee, named team members with real experience, a detailed description of the process, or a sample of the actual work. Specificity is the signal buyers respond to, and it does not require testimonials to achieve.

• Name the person, role and organisation on every quote you use
• Prefer concrete detail about the problem solved over general praise
• Link ratings to their source platform so buyers can verify
• Place proof next to the claim it supports, not in an isolated block
• Never publish invented testimonials, logos or results

Removing the menu, and the trade-off it involves

Stripping the header menu from a campaign landing page is a common tactic, and the logic is sound. Every menu item is an exit route paid for with an ad click. Removing them keeps attention on the single decision the page exists to prompt.

It is not automatically right. Higher-consideration purchases often need the visitor to check credibility elsewhere on the site, and blocking that can cost more than the distraction saves. A useful middle position is to keep a logo that links home, drop the rest of the menu, and put the genuinely needed reassurance directly on the page instead. If people leave to check your about page, the answer is usually to bring that reassurance onto the landing page rather than to force them to stay.

Treat it as a hypothesis to test rather than a rule to obey. It is cheap to try both versions on a page with enough traffic, and the answer differs by offer, price point and how well known the brand is.

Testing landing pages without fooling yourself

A landing page is rarely right first time, so plan for iteration. The mistake is running tests that cannot produce a trustworthy answer: too little traffic, too many changes at once, or a result called the moment it looks good.

Decide before launching what you are testing and what would count as a result. Test the biggest levers first, because they move numbers enough to detect. Headline and offer framing usually beat button colour by a wide margin. Change one meaningful thing per test where traffic allows. If traffic is low, test bigger differences between two complete versions rather than fine variations, because small differences need volume you do not have.

Run for full weeks so weekday and weekend behaviour balance out, and stop at the sample you planned instead of the moment one version pulls ahead. Early leads reverse constantly. Also watch what happens after the conversion. A page that doubles form fills but halves qualified enquiries has made things worse, and you only see that if you track the outcome that pays rather than the click that flatters.

• Write the hypothesis and success measure before you launch the test
• Test headline and offer before cosmetic details
• Run full weeks and stop at the planned sample, not at the first good moment
• With low traffic, test bold alternatives rather than small tweaks
• Measure qualified enquiries, not just form submissions