One page, one job
A landing page is built around a single decision. Someone arrived from an ad, an email, a social post or a search listing with a specific expectation, and the page has to carry them from that expectation to one action: request a quote, start a trial, book a call, buy. Everything on the page is judged against whether it moves that decision. A section that does not is not neutral; it is a distraction, and on a page with paid visitors a distraction costs money.
This is what separates a landing page from the rest of the site. The homepage has to serve returning customers, job seekers, journalists and the merely curious. A product page has to serve people who may want a different product. A landing page serves the people one campaign sends and is allowed to ignore everyone else. That permission is what makes it convert: it can speak to one need, remove the navigation that would lead the visitor away, and repeat one request until the visitor acts or leaves. Its conversion rate (CVRShare of visits or clicks that convert.) is the number that says whether it did its job, and the landing pages hub sets out how that number sits alongside the campaign's other measures.
The "one job" rule has a corollary. If a campaign has two jobs, it needs two pages. A page that asks for a demo and also sells a self-serve plan is asking the visitor to make a decision the business should have made for them.
The sections
Most converting pages contain the same parts, in roughly this order. The table names each, the job it does and the question in the visitor's head that it answers. Not every page needs every section; a short, high-intent page may be headline, offer, proof and form. A page for a considered purchase needs all of them.
| Section | Job | The visitor's question | Common failure |
|---|---|---|---|
| Headline | Confirm the visitor is in the right place and name the outcome | "Is this what I clicked for?" | Clever instead of clear; the brand's slogan rather than the visitor's need |
| Subhead | Add the specific: who it is for, how it works, what is different | "Is this for someone like me?" | Repeats the headline; says nothing new |
| Hero and offer | Show the thing and state what the visitor gets and on what terms | "What exactly am I getting?" | A stock photo and a vague promise |
| Primary call to action | Ask for the action, early, in words that describe the result | "What do I do now?" | Appears only at the bottom; says "Submit" |
| Benefits | Translate features into outcomes the visitor cares about | "Why does this matter to me?" | A feature list copied from the product sheet |
| How it works | Reduce uncertainty about the process after the click | "What happens next?" | Missing entirely on service pages |
| Proof | Give reasons to believe: reviews, logos, results, credentials | "Why should I believe you?" | Anonymous quotes; logos nobody recognises |
| Objections | Answer the reasons people hesitate, in their own words | "What is the catch?" | Hidden in a FAQ the visitor never reaches |
| Guarantee or risk reversal | Remove the cost of being wrong | "What if it does not work out?" | Buried in terms and conditions |
| Form or final call to action | Capture the action with the fewest fields the next step needs | "Is this worth the effort?" | Asks for everything sales might ever want |
| Footer | Legal, privacy, contact; nothing that leads away | "Is this a real company?" | Full site navigation reappears here |
The two sections most often done badly, proof and the form, have guides of their own: the trust and proof guide and the forms guide. This page covers how they sit within the whole. When a new page is commissioned, a landing page brief lists these sections in order with the job of each, so the designer builds from decisions rather than defaults.
Headline, subhead and offer
The headline is read by nearly everyone and is the only section that is. Its job is recognition before persuasion: the visitor should see, within a second, that the page is about the thing they clicked for. The simplest way to achieve that is to use the words of the ad or the query, or their plain equivalent, in the headline itself. That is message match, and the message match guide covers how to keep the ad, the keyword and the headline in agreement across many campaigns. Here the point is narrower: a headline that is clever, abstract or brand-led fails the recognition test even if it is a good line.
The subhead carries the specific. Where the headline names the outcome, the subhead says for whom, how, and what is different about this version of it. "Boiler replacement in two days, fitted by our own engineers, with a long guarantee" is a subhead; "We care about your comfort" is not. A useful test is whether the subhead could be true of a competitor; if it could, it is not yet specific enough.
The offer is the thing itself: what the visitor gets, on what terms, and what happens when they act. It should be stated, not implied. If the page is a lead form, the offer is the call, the quote or the audit that follows, and the page should say what that involves and how soon. If it is a sale, the price or the way to get it should be visible. Hiding the terms raises the conversion rate of the form and lowers the quality of what comes through it, which moves the problem to the sales team.
The hero image or video supports the offer; it does not replace it. Show the product in use, the result, or the people who deliver the service. Decorative imagery takes the most valuable space on the page and gives nothing back. The writing for the web guide covers the craft of the sentences themselves: short, front-loaded, scannable.
Benefits, proof and objections
Benefits are features translated into consequences. A feature is what the thing has or does; a benefit is what changes for the visitor as a result. "Same-day dispatch" is a feature; "ordered by noon, with you tomorrow" is the benefit. The translation is done by asking "so what?" of each feature until the answer is something the visitor would say they wanted. Three to five benefits, each with a short heading and a sentence or two, is more persuasive than a dozen bullet points, because the reader can hold a few things in mind and will not read twelve. The features, advantages and benefits (FAB) chain sets out that translation step by step, with the advantage as the link that makes each benefit believable.
Proof is what makes benefits believable. Every claim the page makes is a claim the visitor has heard before from someone else, so the page has to show evidence: attributed reviews, recognisable customer logos, specific results, credentials, press, awards. Proof works best next to the claim it supports rather than collected in a block of its own, which is why the sections on this page interleave. The trust and proof guide covers the kinds of proof, their placement and the honesty rules; the point here is that a benefits section without proof beside it is an assertion.
Objections are the reasons people do not act. They are specific to the product and are discovered, not invented: sales calls, support tickets, chat transcripts, review sites and on-page surveys say what people worry about. Price, time, effort, risk, trust and "will this work for my case?" cover most of them. The page answers them where they arise, in the visitor's own language, and does not wait for a FAQ at the bottom. A short "how it works" section answers the process objection; a guarantee answers the risk objection; a line about who the service is not for answers the fit objection and makes everything else more credible. The tutorial on turning customer language into page copy is the procedure for finding those objections and phrases in reviews, tickets and chats and placing each in the section that answers it.
Rebuilding a page for a payroll service
A payroll provider's landing page for its "small business payroll" search campaign has a headline that reads "Payroll, perfected", a stock photo of a smiling team, six feature bullets and a form asking for twelve fields. It converts 1.4% of 2,800 monthly clicks into demo requests. The rebuild changes the headline to "Small business payroll from £19 a month, set up in a day", adds a subhead naming the three things the service handles, cuts the benefits to four with a sentence each, places two attributed reviews from businesses of the same size next to the benefits, adds a three-step "how it works" strip and a line stating the first month can be cancelled free, and cuts the form to five fields. Over the next two months the same traffic produces a demo-request rate of 3.9%, and the sales team reports that fewer of the requests come from businesses too large for the product.
The call to action
The call to action is the request. It appears early, in the first screen for visitors who arrive ready, and again at each point where a decision has been made easier: after the benefits, after the proof, after the objections. One request, repeated, is not the same as several requests. The failure is a page with "Book a demo", "Download the guide", "See pricing" and "Contact us" all competing; each extra choice lowers the share of visitors who make any.
The button's words describe the result, not the mechanism. "Get my quote", "Start free trial", "Book a call" tell the visitor what happens; "Submit" and "Click here" do not. The sentence next to the button sets expectations: how long the form takes, what happens after, whether a card is needed. Nielsen Norman Group's recommendations on form usability apply here as much as to the fields themselves — the request is part of the form, and ambiguity about what the button does is a reason to hesitate.
Visual prominence is simple: the button should be the most noticeable element on the screen where it appears, in a colour used for nothing else, large enough to tap, with space around it. A secondary action, if the page genuinely needs one, is styled as a link rather than a second button, so the hierarchy is clear.
The call to action is also where the key performance indicator (KPIA metric tied directly to an objective.) of the page is defined. The event that fires when the action completes is the page's conversion, and it should fire once per completion, carry the campaign's parameters and match what the business counts as a result. A button that leads to a second page where the real action happens means the measurement, and the page's design, should follow the visitor there.
Order by intent
The sections are the same; the order is not. A visitor's readiness when they land decides how much the page has to do before it asks. The See-Think-Do-Care framing names the stages: someone in Do has decided and needs the shortest path; someone in Think is comparing and needs reasons and proof before the request; someone in See has only just noticed the category and needs education the landing page is probably the wrong place to give.
For high-intent traffic — brand searches, searches naming the product, returning visitors from email — the page leads with the offer and the form, puts proof beside the form, and keeps benefits and objections short and below. Making a ready visitor scroll past persuasion they do not need loses some of them.
For comparison-stage traffic — generic category searches, paid social to a cold audience, display — the page leads with recognition and the strongest benefit, builds through proof and objection handling, and makes the main request after the case has been made, with a lighter request (a guide, a price list, a short quiz) available higher up for those who want it.
The click-through rate (CTRShare of impressions that result in a click.) of the ad and the page's conversion rate read together tell you which page you have. A high click-through with a low conversion on a page that leads with the form suggests the traffic was less ready than assumed; the same pattern on a page that leads with persuasion suggests the page is making ready visitors wait.
Reading the order is a research task before it is a design task. Session recordings, scroll maps and the search terms that bring traffic say what people wanted and where they stopped, and the research section of the CRO hub covers those methods. Whatever order the sections take, they have to load quickly and read well on a phone, which the speed and mobile guide covers.
To check an existing page against these sections systematically, use the landing page audit, which walks through message, proof, form and speed so weak points can be ranked before any redesign.
For pages whose goal is an enquiry, the lead form conversion rate tutorial shows how to clean the lead count and measure the rate for one traffic source at a time.
If a page built on these sections still loses most visitors at once, the page on a high landing page bounce rate separates first-screen problems from slow loading, low-intent traffic and overlays.
A popup competes with the page's one job, so it is worth testing before it is adopted; the exit-intent popup test judges it on revenue per visitor rather than sign-ups.
Whether a paid social campaign needs a landing page at all, or a native form inside the platform would do, is covered in the comparison of native lead gen forms and landing pages.
Common mistakes
- A clever headline. Recognition first; wit costs a second the visitor may not give.
- The brand's story before the visitor's need. Who the company is belongs after what the visitor gets.
- Features without translation. A spec sheet is not a benefits section.
- Proof collected in one block far from the claims. Put evidence beside the thing it proves.
- Objections hidden in a FAQ. Answer the reasons to hesitate where they arise.
- Several calls to action. One request, repeated, in words that describe the result.
- The same order for every traffic source. Ready visitors want the form; comparing visitors want the case.
- Site navigation restored. Every link away is an exit the campaign paid for.
- Decorative hero imagery. The most valuable space on the page should show the thing or the result.
