Let’s talk

Web & mobile · Landing pages

Landing pages that convert: one offer, a clear first screen and tests that mean something

Most landing pages lose visitors for ordinary reasons: two offers fighting over one button, a first screen about the company instead of the deal, a form that asks for a phone number nobody plans to call. Fix those first. Then run tests whose winner is still the winner a month later.

One page, one offer

A landing page has one job: get a visitor who arrived with a specific expectation to take one specific action. Everything on the page either helps with that action or competes with it.

Match the ad. If the ad promised “free installation in Houston this month,” the headline repeats that promise in the same words. A visitor who clicked on one offer and lands on a general page has to search for it, and most won’t.

One audience per page. If you run separate ad groups for dentists, clinics and labs, give each its own page, or at least its own headline, proof and form copy. A page written for everyone persuades no one in particular.

One primary action. Remove the main navigation; every link off the page is an exit. A secondary action is fine if it serves the same decision — “download the spec sheet” next to “request a quote” — but not “read our blog” or “follow us.”

The first screen does most of the work

On a phone, the first screen is the only part of the page every visitor sees. It has to answer four questions without scrolling: what is it, who is it for, what do I get, what do I do next.

  • Headline: the outcome in the customer’s words. “Same-week AC repair, flat $89 diagnostic” beats “Your trusted HVAC partner.”
  • Subhead: the specifics that make the headline believable — area served, timing, what’s included.
  • Button: a verb and an object. “Get my quote,” not “Submit.”
  • Visual: the product or the result in use, not a stock photo of a handshake.

Check it on a mid-range phone, not on the designer’s 27-inch monitor. Then try a five-second test: show the first screen to someone outside the project for five seconds and ask what is being offered. If they can’t say, the headline isn’t done.

Proof and form friction

Proof works when it is specific and sits next to the claim it supports. A review with a name, a role and a detail (“they replaced the compressor on a Saturday”) does more than five stars and “great service.” Client logos help only if this audience recognizes them. Put the strongest proof near the form, where doubt is highest.

Answer objections on the page. Pull the questions your sales team hears on every call — price, timing, contract length, what happens after I sign up — and answer them in a short FAQ beside or below the form.

Every form field has to earn its place. Ask only for what you need for the next step. If nobody will call, don’t ask for a phone number. If you need more detail, use a two-step form that opens with the easy question (“What type of property?”) and asks for contact details second.

The small things add up:

  • the right mobile keyboard for each field (numeric for ZIP codes, email for email);
  • autofill that works;
  • errors shown next to the field, in plain words, without wiping what the visitor typed;
  • one line under the button that says what happens next: “We reply within one business day with a written quote.”

Then submit a test lead yourself, on a phone, and confirm it arrives in the CRM with the campaign source attached. Broken forms are more common than anyone likes to admit, and no A/B test will catch them.

Speed is part of the offer

With paid traffic, you pay for every click, including the visitors who leave before the page appears. Google’s Core Web Vitals treat a Largest Contentful Paint of 2.5 seconds or less as “good”; on a landing page, that largest element is usually the hero image or the headline.

The usual fixes are dull and effective: compress the hero image and serve it at the size the screen needs, skip autoplay video on mobile, load chat widgets and heatmaps after the page is usable, and remove tags left over from old campaigns. Measure on mobile with the field data in PageSpeed Insights, not only the lab score.

A/B tests that mean something

Test big things. The offer, the headline, the form length, the proof. Button color is rarely where the difference is.

Pick one metric before you start. Qualified leads or purchases, not clicks. A variant that gets more form fills but fewer sales is a loser.

Calculate the sample size first. On a page converting at 3%, reliably detecting a lift to 3.6% takes roughly 14,000 visitors per variant at the usual 95% confidence and 80% power. If the page gets 2,000 visitors a month, that test runs for over a year. Test something bolder, or use qualitative methods instead: session recordings, five-second tests, calls with recent customers.

Run full weeks and don’t peek. Weekday and weekend visitors behave differently. Stopping the moment a variant looks ahead is the most common way to crown a false winner. Keep traffic sources and budgets stable while the test runs.

Keep a log. Hypothesis, dates, traffic, result, decision. Six months later it is the only thing that stops the team from running the same test again.

What to do this week

  1. Put your top ad next to its landing page and check that the headline repeats the promise.
  2. Open the page on a mid-range phone over cellular and note what the first screen says and how long it takes to appear.
  3. Cut one form field and add a line about what happens after the visitor submits.
  4. Submit a test lead and confirm it lands in the CRM with its source.
  5. Before the next test, calculate the sample size. If it is more than three months of traffic, test something bigger.

Boson Group builds landing pages and runs the performance marketing that sends traffic to them, so the ad, the page and the test plan can be worked out together. If you have a page that gets clicks but not leads, write to team@boson.group.

More about Web & mobile at Boson Group

Contact

Send a brief — we reply with a time and cost estimate within 24 hours.

Brief

Tell us about the project.

Send a brief — we reply with a time and cost estimate within 24 hours.

What do you need?

* required