How to structure a landing page that converts

Most landing pages are built from the inside out: the supplier’s services in the supplier’s order, under a headline that summarizes the company for people who already know it. A stranger reads the page in a different order, and that order can be written down. This article describes an eight-question ladder, checked against published research on how business buyers choose a supplier, and the questions your business has to answer to fill each section.

Why does the order matter?

Because a question lower on the list is inaudible until the ones above it are answered. A price on the first screen is wasted: the reader has yet to decide whether the page is about them. A case study before the problem is a number about a company the reader does not yet care about. So the order is the sequence in which a stranger can take information in, and style and layout follow from it.

Julian Shapiro’s test for a header is whether a visitor who reads nothing else knows exactly what is sold. The subheader then says how the claim is possible and why it is believable. Nielsen Norman Group’s B2B usability research finds that a site has to establish credibility before a visitor will hand over contact details. That is why proof comes before price, and price before the button. Gartner’s six buying jobs describe the same thing from the buyer’s side: problem identification, solution exploration, requirements building, supplier selection, validation and consensus creation. Gartner adds that buyers loop through the jobs out of order. The page does the buyer’s jobs in the buyer’s sequence, and a reader who jumps back finds the section where they left off.

What does the buyer research say?

April Dunford’s positioning method starts from the competitive alternatives, and the first alternative is always the status quo: what the customer does today if the supplier did not exist. A problem stated the way the reader lives it is how a page names the status quo. B.J. Fogg’s Stanford guidelines for web credibility define credibility as perceived expertise plus perceived trustworthiness, and the first guideline is to make it easy to verify the accuracy of the information. The second and fourth are to show that a real organization and honest people stand behind the site.

Hinge Research Institute’s study of how buyers buy professional services puts reputation, then expertise and specialization, at the top of the selection criteria, with price deciding roughly 8 percent of the time. That study covered 822 buyers and is now around a decade old, so the figure is directional. Matthew Dixon and Ted McKenna’s analysis in Harvard Business Review in 2022, built on 2.5 million sales conversations, found 40 to 60 percent of qualified deals lost to no decision at all. The remedy they propose is to take risk off the table, offer a recommendation and limit the options.

Marcus Sheridan’s five topics buyers research before buying are cost, problems, comparisons, reviews and who is best; two of the five are comparative. Bain’s B2B elements of value, published in Harvard Business Review in 2018, count risk reduction, reduced anxiety and reputational assurance among the elements that predict loyalty. Google’s research on the messy middle lists social proof and authority among the six biases that move a decision. Taken together, the research says a page should name the problem, prove the expertise, compare honestly, show the price and take risk away. The ladder below is those five things in the order they can be heard.

1. What is this, what do you do?

The section holds one sentence a reader can repeat to a colleague: what the company makes and how. The headline is the answer. The section fails when the reader has to assemble the answer from an eyebrow, a headline and a lede, or when the headline names an audience and so turns every other reader away. Three questions fill the section.

• How would you describe the business to a stranger in one sentence, in your own words?
• A headline carries one thing first: which is it?
• Does the size or shape of the team belong on the first screen, or later?

2. What problems do you solve?

The section holds the problems in the reader’s words, each paired with what you do about it and a link to the service that does it. This is where the reader recognizes themselves, and it is why a headline naming a role is redundant: a person who sees their own situation described already knows the page is for them. The section fails when it lists services by name, sorts readers by title, or describes the problem the way the supplier sees it. Three questions fill the section.

• What have customers actually said when they first got in touch, in their words, per service?
• Which problems have people asked you about in the last three months, ranked?
• What do you get asked for that you would prefer to stop attracting?

3. How do you work, and what do I get?

The section holds the process with durations: call, proposal, build, handover. Then what lands at the end — what the customer owns when the work is done, and where it lives. It also says how quality holds, as a mechanism the reader could go and check. The section fails when it reads as a values statement; “we care about quality” can neither be checked nor disproved. One example of durations published per phase is the article on how long a website takes. Three questions fill the section.

• Are the steps and durations you publish still accurate?
• What does the customer have to put in: hours per week, which people, what material?
• What exactly arrives at handover, and what support follows it?

4. Why should I believe you, and who is behind this?

The section holds real proof, each piece with one outcome a buyer understands at once, and your relationship to the client beside it. Under the proof sits the organization, stated plainly: country, registration number, year, and who does the work. This is Fogg’s second guideline in practice, and it is also the part of the page a small supplier with few named clients and no logos can still fill honestly. The section fails when the figures are engineering figures, when the proof sits below the price, or when the supplier is a name with nothing behind it. Three questions fill the section.

• Legal name, registration number, year founded, location?
• Which three pieces of proof lead (cases, results, references), and what is the one sentence a buyer should take from each?
• What can you point to outside your own site: a public repository, a published tool, a talk, years of experience stated as the organization’s?

5. Why you, and when is someone else better?

The section holds the alternatives on the reader’s shortlist, including keeping things as they are, and when each one wins. Your own column says the same about you. Dunford’s point is that a capability is only differentiated relative to the alternatives actually under consideration, and Sheridan’s comparisons topic is the same question asked by the buyer. The section fails when one column wins every row. That is an ad, and both readers and the systems that cite pages treat it as one. Four questions fill the section.

• Who do you actually lose to: an agency, a freelancer, a hire, in-house tools, an off-the-shelf product, doing nothing?
• When is each of them honestly the better choice?
• Which of your differences would a buyer actually care about?
• What do you not do?

6. What does it cost, and how long?

The section is one line: the price bracket, the timeline, and when the price becomes fixed, with a link to the pricing page. The research pulls two ways here and both are right. Nielsen Norman Group ranks pricing as the information business buyers want most, and Sheridan’s observation is that a supplier who hides the price is assumed to be hiding something. Hinge’s 8 percent says price rarely decides. So price is a qualifier the reader wants early and an argument that wins late. It never leads the first screen, it never hides behind a form, and the pricing page is one click away in the navigation at all times. One example of a published ladder, with what each tier includes, is the website pricing page. The section fails when price leads the first screen, hides behind a form, or appears more than once on the page. Two questions fill the section.

• Are the published figures right, and are they at or above the market floor?
• Does the page state a floor, or the range most projects land in?

7. What if it goes wrong, and what happens next?

The section puts risk beside the ask: the terms you stand behind when something goes wrong, who owns the result, and what happens if the customer wants to switch supplier. Then the commitments: how fast a reply comes, how fast a written proposal follows, and that the customer decides at their own pace. This is Dixon and McKenna’s strongest lever, and Bain’s risk reduction and reduced anxiety, placed where the decision is made. What ownership means in practice is in the article on owning your website. The section fails when the guarantee is a footnote, or when the button promises something the page it leads to does not do. Four questions fill the section.

• What terms would you stand behind: a fixed price with overruns carried by you, a guarantee, a delivery-date promise, ownership of the result?
• Does the button do what it says? A “book a call” that opens a form is a broken promise.
• Reply time and proposal time, and on which days?
• Who replies, and how?

8. What am I still unsure about?

The section holds four to eight questions the reader has at this point, answered plainly. The set belongs to this page and no other, because identical FAQ markup on a dozen URLs tells a search engine those pages are interchangeable. After the questions come a few articles as the expertise signal, since Hinge’s research finds that buyers judge expertise from content: case studies, specialized articles and client lists. The section fails when an answer contradicts the offer the rest of the page describes. Three questions fill the section.

• What questions come up in first calls and emails, raw?
• Which objections do you hear most?
• For a small supplier: what happens if the key person is unavailable mid-project?

Which rules hold across the whole page?

Seven rules hold the ladder together, and they apply to every landing page whatever it sells.

• One question per section, and the heading is the answer. “Our services” is a label; “We shorten the time from order to invoice” is an answer. A reader skimming headings alone should get the whole argument.
• Problems in the reader’s words, method in the supplier’s. Question 2 is written the way the reader would describe the situation to a colleague. Question 3 is written the way you would explain it on the call.
• Proof a buyer can read. Every outcome carries its plain meaning; the technical figure belongs on the case page.
• Price once, and never leading. One line in question 6, and the pricing page always one click away.
• Every alternative gets its honest case. Question 5 names when an agency, a freelancer, a hire, a site builder or doing nothing is the better choice.
• The page is forwardable. Every heading plus its first line survives being pasted into an email to a colleague, and every figure on the page matches the price list and the written proposal exactly. Gartner’s sixth buying job is consensus creation, and the reader has to be able to pass the page along.
• Nothing is written for the section alone. Every sentence is content the site already publishes: the services, the proof, the About page, the FAQ sets. A landing page assembles what the site already says.

How does the same ladder apply to other page types?

The order never changes. What changes is where the reader enters and how narrow each answer is. On the home page the reader enters at question 1 and the ladder covers the whole offer, with one problem per service in question 2. On a service page the reader also enters at question 1, but the ladder covers one service: two or three problems, and that service’s own alternatives in question 5. Its price ladder sits high, because the searcher who arrived there asked about price.

On a pricing page the reader enters at question 6, so the page opens on the ladder, then proof, then the comparison, then risk. A case page opens at question 4: the outcome first, then the relationship and the limits, then how it was built and what happens next. An article opens at question 8, on one question the reader typed into a search box. The direct answer opens the page and the offer appears once, at the end, as question 7.

Where is the ladder in use today?

Torn Studio uses the ladder on its own pages. On September 8, 2026, the studio decided to rebuild its home page on it and to drop role-based messaging from its landing pages. The cost was three audience pages and the home page copy in nine languages. The same ladder sits under the websites the studio builds, where question 2 becomes the problems the site has to solve and question 6 becomes a fixed price before work starts.