Landing page redesign: a step-by-step process

What is a landing page redesign?

A landing page redesign is a rebuild of a page’s message, structure and design, aimed at turning more of the visitors you already pay for into customers. It is different from a visual refresh. A refresh changes how the page looks. A redesign changes what the page says, in what order, and what it asks the visitor to do.

I use the same eight steps every time. The first two are research, the middle four are design and build, and the last two decide whether the new page actually earned its place.

What are the steps of a landing page redesign?

Eight steps, in this order.

1. Audit the current page and its data.
2. Find the leaks.
3. Rewrite the first screen to continue the ad.
4. Structure the page: promise, proof, objections, offer, call to action.
5. Design mobile-first.
6. Build it.
7. A/B test it against the old page.
8. Read the result in revenue.

How do you audit the current landing page?

Open the ad, the page on a real phone and the page’s analytics side by side, and write down what you see before you change anything. I look at where the traffic comes from, the conversion rate and revenue per visitor by device, how far people scroll, and where they drop out of the form or checkout. Session recordings and a short on-page survey help explain the numbers.

I also check tracking before trusting any of it. If the conversion fires twice on desktop or not at all on mobile, the old page will look better or worse than it is, and so will the test later. My CRO audit checklist covers the full set of checks.

How do you find where the page leaks visitors?

Look for the biggest drop between two steps, then ask what on the page causes it. Common leaks on paid landing pages:

• The first screen. Many visitors leave without scrolling or clicking. Usually a message match problem.
• The call to action. People scroll but don’t click. Usually weak proof, an unclear offer or a vague button.
• The form or checkout. People start but don’t finish. Usually too many fields, surprise costs or errors.
• Mobile only. Desktop converts and mobile doesn’t. Usually layout, speed or tap targets.
• Exits through navigation. Clicks on the menu, logo or footer links that never come back.

Write each leak as a sentence: “Most mobile visitors leave on the first screen, and the first screen doesn’t mention the ad’s promise.” That sentence becomes the brief for the redesign.

How should the first screen change?

Rewrite it so it continues the ad: the same promise in the headline, the same visual, and a first action that is the natural next step. The visitor clicked for a specific reason. The first screen should confirm that reason within a few seconds. A brand slogan or a “Welcome” headline makes them work it out. See ad-to-landing page message match and what to put above the fold for the details.

How should a redesigned landing page be structured?

Follow the questions a visitor asks, in the order they ask them: promise, proof, objections, offer, call to action.

• Promise: “Is this what I clicked on?” The headline, subhead and visual.
• Proof: “Does it work, and do other people trust it?” Ratings, reviews, numbers you can stand behind.
• Objections: “What could go wrong?” Price, side effects, time, commitment, returns. Answer them plainly.
• Offer: “What exactly do I get, and for how much?” The product, price, bundle or guarantee.
• Call to action: “What do I do now?” One action, repeated, with a button that names the outcome.

Here is how the sections of a typical old page change in a redesign.

Why design the landing page mobile-first?

Because most paid social traffic arrives on a phone, and a layout that works on a small screen is easy to widen, while the reverse rarely works. On the phone, I make sure the headline, the first action and one line of proof fit on the first screen, buttons are thumb-sized, and nothing covers the call to action. Then I check speed: Google’s “good” mark for Largest Contentful Paint is 2.5 seconds or less, measured at the 75th percentile of page loads.

What matters when building the new page?

Keep the tracking identical to the old page, keep it fast, and make it easy to change one section at a time later. The same conversion events, with the same values, must fire on both pages, or the A/B test compares two different rulers. I build sections as reusable blocks so later tests (a new headline, a different proof block) don’t need a rebuild. I also test the full path on a real phone and a desktop before any traffic goes to it.

How do you A/B test a redesign against the old page?

Split the traffic from the same ads evenly between the old page and the new one, and run it long enough to get a clear answer.

• Primary metric: revenue per visitor (value per session).
• Guardrails: order or lead quality, refund rate, average order value.
• Duration: at least two full weeks, and until the result is statistically clear.
• Same traffic: both pages get the same ads, audiences and devices during the same dates.

A full redesign changes many things at once, so a win tells you the new page is better, not which part did it. That is fine for the first test. After that, I test single sections on the winning page to learn what carries the result.

How do you read the result in revenue?

Multiply the difference in revenue per visitor by the number of visitors who will see the page in a year, using a conservative estimate. Conversion rate alone can mislead: a page can get more orders at a lower order value, or more leads of lower quality. Revenue per visitor captures both. The full method is in how to calculate the revenue impact of an A/B test.

If you want a page redesigned and tested this way, see landing page design or book a call.

FAQ

Sources

• web.dev: Largest Contentful Paint (LCP)
• Google Ads Help: About Quality Score for Search campaigns
• CRO audit checklist: 25 checks for a landing page