Landing Pages That Convert: The Working Guide
A landing page converts when it keeps the promise that brought the visitor, states one offer plainly, and asks for one action. The bench guide to building pages that do those three things at small-business scale.
The short answer
A landing page is any page a campaign sends traffic to, and it converts when three things line up: it keeps the exact promise the ad, email or link made (message match), it presents one offer with one clear next action, and it removes everything that competes with that action. Landing pages fail in predictable ways - the headline talks about the company instead of the offer, the page asks for three things at once, or the form demands more than the visitor's interest can pay for. This bench works through each element in its own guide; this pillar is the map.
Start from the promise, not the page
Every landing visit begins before the click. The visitor read an ad headline, an email subject, a search result - and formed an expectation. The page's first job is to confirm that expectation in the visitor's own words within seconds, because the alternative reading is 'wrong page' and the back button is right there. Nielsen Norman Group's work on how long users stay on web pages describes the window bluntly: pages have roughly 10 to 20 seconds to make their case, and most of that judgement happens at a glance. Message match is not a copywriting nicety; it is the mechanism by which the click you paid for becomes a visit at all.
One page, one job
The second discipline is singleness. A landing page for a spring sale on walking boots should sell walking boots in the spring sale - not also recruit newsletter subscribers, promote the loyalty scheme and introduce the brand story. Secondary asks are not free: each one divides attention and gives the visitor a cheaper action than the one you want. Stores sometimes fear that a focused page wastes the traffic that does not convert; the follow-up bench exists precisely to catch those visitors by email with their consent, not by cluttering the page they are currently deciding on.
The elements, in the order visitors check them
Visitors read a landing page in a rough order: headline and lead image (is this what I clicked?), the offer's terms (price, what is included, what it costs to get it here), proof (can I believe this seller?), and the action (what exactly happens when I tap?). The bench mirrors that order: headlines, the first screen, trust elements, and the form or buy action. Work in that order and you will fix causes before symptoms.
The traffic source rewrites the page's job
A cold visitor from a social ad needs the offer explained; a warm visitor from your own email list needs it confirmed and made easy. Search traffic arrives with a question phrased in their words. One page can serve one of these audiences well. This is why the bench has separate guides for email traffic and search traffic, and why 'the landing page converts at X' is a meaningless sentence until the source is named. Build per audience when the volume justifies it; otherwise, write for the source that sends the most valuable traffic and let the others be served adequately.
Measure the page against its own promise
A landing page's conversion rate is conversions divided by the sessions that arrived from its campaign - not store traffic in general. Judge changes against the page's own baseline for the same source, over whole weeks, and distrust any read that mixes a new audience into the comparison. Most landing page 'failures' this bench sees are measurement failures underneath: the page was fine, the traffic was different. The foundations bench covers the baseline discipline; this bench assumes it and spends its words on the pages themselves.
The shape of a working campaign page
Illustration, assembled from the bench's guides in order. First screen: the offer headline echoing the ad, the product visual, the terms line (price, delivery, returns), the action. Second screen: the three doubts answered - proof beside the product, the guarantee beside the price, the delivery detail beside the total. Third screen: specifics for the decided visitor - sizes, compatibility, what is in the box. Foot: the action repeated, the contact route for the hesitant. Nothing on the page competes with the corridor the visitor is walking. Real pages vary in costume; working pages rarely vary in this skeleton, because the skeleton is the visitor's decision sequence wearing page clothes.
Sources and evidence on the record
- Nielsen Norman Group, How long do users stay on web pages? - the short window in which a page has to make its case.
- Unbounce, Conversion Benchmark Report - landing page results analysed across industries; cited here for how widely rates vary between industries, not as a target.
- CXL, CXL blog - practitioner writing on testing method, research process and optimisation programme discipline.
Evidence basis: Method guide. Any figure quoted is attributed to a named published source listed below; worked examples are labelled illustrations with round numbers, not measurements. Applies when: You send any paid, email or social traffic to a specific page