Landing pages that convert: what to put on the page, in order
The parts of a landing page that turn a visitor into a lead or a sale, the order they belong in, and the habits that quietly cost a page its customers.
A landing page has one job, which is why it is usually easier to fix than a whole website. Somebody arrived because of an ad, an email or a search. They have a question in mind and a few seconds of patience. The page either answers the question and offers the next step, or it does not, and most of what makes the difference can be seen in the first screen.
What has to be in the first screen?
- A headline that repeats the promise of the ad or email that sent the visitor, in the same words where possible.
- One line underneath that says what you offer and who it is for.
- A picture of the actual product, service or result, not a stock photo of people smiling at a laptop.
- The one button, saying what happens when it is pressed: Book a call, Get the price list, Order now.
Most visitors on a phone never scroll past the first screen. If they cannot tell what the page is and what to do in that space, the rest of the page is written for nobody.
What order should the rest of the page follow?
- The promise and the one action, as above.
- Proof: real work, real reviews, named clients where you are allowed to name them.
- How it works, in a few plain steps, so the visitor can picture what happens after they press the button.
- What it costs, or at least what it starts from. Hiding the price sends people away to find it somewhere else.
- The doubts, answered: the questions people ask before they buy, each with a short, honest answer.
- The action again, for the visitor who read everything first.
What quietly costs a page its customers?
| Habit | Why it hurts | What to do instead |
|---|---|---|
| Full site navigation at the top | Every link is a way off the page | Logo and the one action only |
| A long form | Each field is a reason to stop | Ask only what you need to reply |
| Several different buttons | The visitor has to choose, and often chooses neither | One action, repeated |
| A headline about you | The visitor came with their own question | A headline about their result |
| Slow loading on a phone | People leave before it appears | Compress images, cut scripts and pop-ups |
| Claims with no proof | Nobody believes them | Show the work or a review |
What does proof look like when you are new?
You do not need a wall of logos. A photo of the real product, a short explanation of how you work, the name and face of the person who will do the job, and a clear guarantee or refund policy all do the work proof does: they make the page believable. What does not work is inventing it. A made-up testimonial or an unearned badge is easy to spot and costs far more trust than an empty space.
We hold our own pages to that rule. We name a client only where the client has agreed to be named, which is why the Gettysburg Museum of History appears on this site and others are described by their trade, and we print a result only when it has been checked. It makes for fewer numbers on the page, and every one that remains can be believed.
How do you know whether a change worked?
Change one thing at a time, and give it long enough to see a difference beyond normal ups and downs: the headline, then the picture, then the form. Watch the share of visitors who take the action, not page views. If you change the headline, the photo and the price on the same day and sales go up, you will never know which one did it, and you may keep the change that hurt. Our pieces on reading an ad account, at /blog/read-an-ad-account-in-ten-minutes, and on site speed, at /blog/website-speed-in-plain-english, cover what to look at before you start changing words.
Does every ad need its own landing page?
Not every ad, but every promise. If two ads say different things, offer different products or speak to different people, each should land on a page that continues its own sentence. Sending every ad to the homepage is the most common reason a good ad looks like it failed: the visitor clicked on one promise and arrived at a page about everything.
This is where a landing page earns its keep over a whole website. It can be built quickly, changed without touching the rest of the site, and matched word for word to the ad that sends people to it. When you write a new ad, write the first screen of its page at the same time, so the two are planned as one conversation.
When is a new landing page the wrong fix?
When the traffic is wrong. A perfect page will not convert visitors who were never going to buy, and if an ad is reaching the wrong people, the fix is the ad. And when the offer is the problem: if nobody wants the thing at that price, a better page only shows them more clearly that they do not want it. Fix the audience and the offer first. When those are right and the page is the weak link, our landing page work is described at /services/landing-pages.
Questions we get asked
Usually not. A landing page made for one action works better without links leading elsewhere. Keep the logo, and let it link home if you must.
As long as it takes to answer the visitor's doubts. A simple, cheap offer needs little. A considered purchase needs proof, steps, price and answers.
Where you can, yes, or at least a starting price. Visitors who cannot find a price often leave to look for one and do not come back.
Only the ones you need to reply. Everything else can be asked on the call or after the sale.