Here's a scenario I see all the time. A business spends good money driving traffic to their site through paid ads, the clicks come in, and almost nobody does anything once they arrive. The ads aren't the problem. The page they land on is.
A landing page has one job: get the visitor to take one specific action. Buy, book, call, sign up. Everything on the page should push toward that, and anything that doesn't is in the way. That sounds obvious, and yet most pages quietly break the rule.
One page, one job
The fastest way to kill a landing page is to ask it to do too much. Your homepage can afford to point in ten directions. A landing page can't. If someone clicked an ad about tax filing, the page should be about tax filing, with one clear next step, not a tour of everything your firm does.
This is called message match, and it matters more than almost anything else. The promise in the ad and the headline on the page should line up so closely the visitor feels they're in the right place within a second. When they don't match, people bounce, and you've paid for nothing.
The pieces that actually matter
- A headline that matches the promise
- The first thing people read should echo why they clicked and state the benefit plainly. Clever beats clear far less often than people hope.
- One clear call to action
- One primary button, repeated as the page gets long, saying exactly what happens next. "Get my free quote," not "Submit."
- Proof you can be trusted
- Reviews, real results, recognizable logos, a number that shows you've done this before. People need a reason to believe you.
- As little friction as possible
- Short forms, no surprise steps, no asking for a phone number you don't need yet. Every extra field costs you conversions.
Speed and mobile are not optional
Most of this traffic is on phones, and a landing page that loads slowly or looks broken on a small screen leaks money. I go deep on this in my guide to website speed and Core Web Vitals, but the short version is simple: if the page isn't fast and clean on a mid-range phone, fix that before you touch anything else.
Where the leaks usually are
When a page isn't converting, the cause is rarely mysterious. It's almost always one of a short list of things, and you can check all of them in half an hour without any tooling beyond your own phone.
Work down this table in order. Each row is something I've found on a real page that was quietly costing the owner money, and each fix is small enough to do the same day you spot it.
| What's wrong | How you spot it | The fix |
|---|---|---|
| The headline doesn't match the ad | Read the ad, then the headline, back to back | Rewrite the headline to repeat the ad's promise almost word for word |
| No clear next step above the fold | Open it on a phone and don't scroll | One button, visible immediately, saying what happens when you press it |
| Form asks for too much | Count the fields | Cut to name, contact, and one question. Ask the rest on the call |
| Nothing that proves you're real | Look for a single verifiable detail | Add a review, a named result, or a photo of the actual person who replies |
| Navigation leads people away | Count the exits in the header | Strip the menu down, or remove it. The page has one job |
| Button says Submit | Read the button out loud | Say the outcome instead. "Get my free quote" beats "Submit" every time |
| Page is slow on mobile data | Load it on a phone off wifi | Compress the images, drop the widgets you don't need |
The first and last rows are the ones people skip, and they're usually the two biggest. A mismatch between ad and page wastes the click before the visitor reads a word, and a slow page wastes it before they see one.
Watch someone use it, just once
Before you spend money on tests, hand your phone to a person who doesn't know the business, point them at the page, and ask what you sell and what they'd do next. Don't help them. Ten seconds of watching somebody hesitate will teach you more than a week of guessing in a meeting, and it costs you nothing but mild embarrassment.
Write for the skimmer, not the reader
Nobody reads a landing page start to finish. They scan. So write for scanning:
- Short paragraphs, one idea each.
- Subheadings that carry the argument on their own, so someone reading only those still gets the point.
- Bullet points for benefits, phrased as what the customer gets, not what you do.
- The important stuff, headline, offer, and a button, visible before anyone scrolls.
Test, don't guess
Even good marketers are wrong about what will work more often than they'd like to admit. The fix is to test instead of argue. Change one thing at a time, the headline, the button text, the image, and see what actually moves the numbers. Small, steady improvements compound into a page that converts far better than the one you launched.
- 1Pick one element to test. One, so you know what caused the change.
- 2Give it enough traffic and time to mean something before you call a winner.
- 3Keep the version that wins, then test the next thing.
- 4Write down what you learn. Patterns across tests teach you about your customers.
A great offer on a mediocre page beats a mediocre offer on a beautiful one. Start with what you're actually offering.
The best landing page in the world won't rescue a weak offer, so get that right first. Then make the page fast, focused, and honest, and keep testing. Do that and the same ad spend that was fizzling starts paying off.
If you're pouring money into ads that aren't converting, the page is the first place I'd look. It's where good web development and sharp content overlap. Send me the details and I'll tell you what I'd change.
Frequently asked questions
It varies a lot by industry and offer, but many landing pages convert somewhere in the low single digits up to around 10 percent. Rather than chase a benchmark, focus on beating your own current rate through steady testing.