How to increase a landing page’s conversion rate.
To increase a landing page’s conversion, first find out at which point people give up and only then touch the page. In most cases, the gain comes from four adjustments: a message that matches the ad, a clear first fold, proof near the button and a shorter form.
Fabrício MartinsFounder of Martin Marketing & Design
In short
- What conversion is: the share of visitors who take the action the page asks for.
- Where to start: at the point where most people give up, not with what is easiest to change.
- What weighs the most: offer and message. Button color comes much later.
- How to test: one change at a time, with the deadline and the criterion defined beforehand.
What conversion rate is
Conversion rate is the share of visitors who do what the page asks: fill in the form, message on WhatsApp, buy, book.
Conversion rate = conversions ÷ visits × 100If 800 people visited the page and 32 submitted the form, the rate is 4%.
The number alone says little. A page at 2% can be great for an expensive sale and poor for a free resource. What matters is the comparison with your own history and with the cost of bringing in each visit.
A simple example shows why it is worth the effort. If the page receives 1,000 ad visits a month and converts 2%, that is 20 leads. At 4%, it is 40 leads on the same budget. Doubling conversion cuts the cost per lead in half, without spending one more real on media.
Before you touch anything: find out where the page loses people
The most common mistake is starting with what is easy to change. Before any adjustment, look at the page as a small funnel, with four steps:
- Arrived. The person opened the page and it loaded.
- Stayed. They got past the first fold and scrolled the page.
- Tried. They clicked the button or started filling in the form.
- Completed. They submitted the form or opened the conversation.
Each lost step points to a different problem:
| Where the person gives up | What it usually is | Where to look |
|---|---|---|
| Leaves without scrolling | Weak promise, slow page or the wrong visitor | Headline, speed and ad targeting |
| Scrolls and does not click | Proof is missing or the offer was not clear | Benefits, testimonials and price |
| Clicks and does not submit | Long form or an intrusive request | Fields, error messages and an alternative way to get in touch |
| Submits and does not become a customer | Unqualified lead or slow follow-up | The ad’s promise and the sales process |
Always start with the step that has the biggest drop. Fixing the form on a page nobody scrolls to the end of does not change the result.
To see these steps, the basic events in analytics are enough: scroll, button click, form start and form submission. Heatmap and session recording tools help you see what the numbers do not show, such as people clicking on something that is not a link.
Make the page continue the ad
Whoever clicks on an ad arrives with an expectation. If the ad talks about a "no-obligation website quote" and the page opens with "Welcome to our company", the person feels they landed in the wrong place and leaves.
The rule is simple: the page headline repeats the ad’s promise, in the same words. The image is the same or from the same family. And the button asks for the action the ad announced.
This also applies to the audience. A page for people who have never heard of the company needs to explain more. A page for people who have already visited the website can go straight to the offer. When a campaign speaks to two audiences, the best path is usually to have two pages.
The first fold decides
The first fold is what appears before scrolling. That is where the person decides whether to stay. It needs four elements, and only those:
- A headline with the benefit. What the person gains, not what the company does. "A full schedule without depending on referrals" says more than "Marketing solutions".
- One supporting line. Who it is for and how it works, in one sentence.
- A quick proof. A review rating, the number of customers or a well-known logo.
- The button. Visible without scrolling, with a clear verb.
What gets in the way: a menu with ten links, a rotating banner, a video that starts by itself and corporate boilerplate. Each of them is an exit or a distraction.
The five-second test: show the first fold to someone from outside for five seconds and ask what the page offers. If the person cannot answer, the problem is in the headline.
Put the proof where the doubt appears
A testimonial hidden at the end of the page does little work. Proof pays off more when it sits next to the doubt it resolves.
- Near the button. A short testimonial or the review rating right next to the form.
- After the price. A concrete customer result, to justify the investment.
- Next to each benefit. A number or an example that shows the benefit happening.
Strong proof is specific. "Great service" convinces nobody. "The website was ready before the opening and quote requests started in the first week" does. Name, role and company increase trust, always with permission from the person quoted.
Form: every field costs submissions
The form is where most pages lose people who were already convinced. For each field, ask the question: do I need this now, or can I ask later, in the conversation?
- Cut what is not essential. A name and one way to get in touch solve most cases. Job title, company tax ID and address can wait.
- Keep the label always visible. Example text inside the field disappears when the person starts typing.
- Use the right keyboard on mobile. A phone field opens the numeric keyboard, an email field opens the keyboard with the at sign.
- Explain the error. "Invalid field" does not help. "The area code is missing" does.
- Say what happens next. One line below the button: who replies and through which channel.
- Offer an alternative path. Many people prefer WhatsApp to a form.
A shorter form brings more leads, but not always better leads. If the sales team complains about quality, one more qualifying field can pay off. The decision comes from the data, not from preference.
Speed and mobile
Most ad visits arrive on a phone, often on a mobile network. A page that is slow to open loses people before showing a single argument.
- Light images. At the size they are displayed and in current formats.
- Fewer scripts. Every third-party tool installed weighs on loading.
- Button within reach of the thumb. Big enough and with space around it.
- Text readable without zooming. And with enough contrast to read outdoors.
- Nothing covering the content. Notices and pop-ups that take over the whole screen on mobile drive people away.
Test the page on your own phone, with mobile data on, and try to convert. Five minutes walking the customer’s path reveal more than an hour looking at the computer.
Answer the objections before the button
Every offer has three or four doubts that stall the decision. They arrive every day by phone and on WhatsApp, and rarely appear on the page.
Ask whoever handles inquiries what the most common questions are before a customer closes. They are usually variations of:
- How much does it cost, or from how much?
- How long until it is ready?
- Does it work for my case?
- What if I do not like it?
- How does payment work?
Answer each one with a direct sentence, in a frequently asked questions section right before the last button. Hiding the price rarely helps: whoever cannot afford it becomes a lead the team will waste time on.
How to test without fooling yourself
Everybody has an opinion about a page. The A/B test settles the discussion: half of the visitors see the current version, half see the new one, and the numbers decide.
- Write the hypothesis. If we change this, conversion goes up, for this reason.
- Change only one thing. With two changes together, you never find out which one worked.
- Set the deadline beforehand. At least two full weeks, to cover weekdays and weekends.
- Wait until you have volume. With a few dozen conversions, the difference may be chance.
- Record the result. Including when the new version loses.
The order of what to test follows the impact:
| Priority | What to test | Why |
|---|---|---|
| 1 | Offer | What the person receives weighs more than the way it is presented |
| 2 | Headline and first fold | It is what every visitor sees |
| 3 | Form | It is where people who have already decided give up |
| 4 | Proof and section order | They affect the trust of people who are undecided |
| 5 | Button text and color | A small gain, only worth it with a lot of traffic |
A page with little traffic does not need to give up on improving. Instead of an A/B test, make big changes, one at a time, and compare equal periods. It is less precise, but better than not measuring.
Quick conversion checklist
The page has
- A headline with the same promise as the ad.
- A button visible without scrolling.
- A single goal, with no exit menu.
- Proof next to the form.
- Answers to the most common questions.
- Conversion events being measured.
The page does not have
- A form with fields nobody uses.
- A rotating banner at the top.
- A video that plays by itself.
- Links to social media along the way.
- A pop-up that covers the screen on mobile.
To see each part of one of these pages in place, see the interactive anatomy on our page about landing page design.
Conversion does not end at the button
A lead only becomes a sale if someone replies. A good part of a landing page’s result is lost after the submission, outside the page.
- Thank-you page. It confirms that it worked and states the next step. It is also where the conversion is measured.
- Immediate notice to the team. The lead needs to reach whoever handles it right away, with the campaign it came from.
- An agreed first reply. Who replies, through which channel and with which message.
- A record of the outcome. Whether it became a sale or not, and why.
This record closes the loop: it shows which campaigns bring customers, and not just leads. To do that math, see how to calculate ROAS. And if your question comes earlier, about the right format, read website vs. landing page.
Frequently asked questions
There is no number that fits everyone. The rate changes with the price of the offer, the traffic source and what the page asks for. The best reference is your own history: today’s page needs to convert more than last month’s, at a cost per lead that fits your margin.
It helps when it explains something hard to show in text and when the person chooses to watch. It hurts when it plays by itself, weighs on loading or replaces the headline. The page needs to work for people who never press play.
Whenever the person may be ready to decide: in the first fold, after the benefits, after the proof and at the end. The text and the destination are always the same. Repeating the button is not being pushy, it is sparing the person from searching.
They can increase the number of leads and worsen the experience at the same time. If you use one, prefer the kind that appears when the person shows they are about to leave, with an offer different from the page’s, and never covering the whole screen on mobile right on arrival.
For a reliable A/B test, yes: a few hundred conversions per version give confidence. With less traffic, make bigger changes, one at a time, and compare equivalent periods. The result is an indication, not proof.
Conclusion
Increasing a landing page’s conversion is less about tricks and more about order: finding out where the page loses people, fixing the message and the offer, making submission easier and only then testing the details.
A page that converts twice as much is worth more than twice the budget. If you want a page built this way from the first draft, get to know Martin’s landing page design.
