Most advice about landing pages assumes you haven't built one yet. This is for the other situation: the page is live, people are landing on it, and almost nobody is converting. Work through these five causes in order — they're sequenced so that each one rules out a whole class of fixes below it, and skipping ahead is how founders end up rewriting a headline when the real problem was the traffic.
1. The wrong people are arriving
Before you change a single word, check whether the page ever had a chance.
A page converts a visitor who already wanted the outcome you're offering. If your traffic is coming from a broad social post, a directory listing, or a keyword that means something different than you assumed, then no amount of copywriting will rescue it. You are not looking at a conversion problem — you are looking at a targeting problem wearing a conversion problem's clothes.
How to check: segment by source. If one source converts at a normal rate and the others are near zero, the page works and the traffic doesn't. Fix the traffic, not the page.
2. A visitor can't tell what this is within five seconds
Open your page in a private window and read only what's visible before you scroll. Then ask a specific question: could a stranger say what this does, who it's for, and what happens next?
The failure here is rarely a "bad" headline. It's usually one of three things:
- Clever instead of clear. "Ship faster, dream bigger" describes nothing. Being boring and specific beats being memorable and vague, every time.
- Feature-first instead of outcome-first. Visitors buy the result. The feature list is the proof, not the pitch.
- Written for you, not for them. Internal vocabulary — your product's coined terms, your category name — is invisible to you and opaque to everyone else.
If a page fails this test, nothing further down the page matters, because most visitors never see it.
3. There's no reason to believe you
A stranger's default assumption is that your claim is marketing. Something has to close that gap, and it has to be visible without scrolling far.
What actually works: a real customer's words with a real name, a specific number you can substantiate, a recognizable logo you have permission to use, or a live demo that lets someone verify the claim themselves.
What doesn't: stock photos of people who aren't your customers, unattributed praise, and vague badges. These don't read as neutral — they read as evidence that you had nothing better, which is worse than showing nothing at all.
If you're pre-launch and have no proof yet, say so plainly. "We're new — here's exactly what we've built so far" converts better than manufactured credibility, because visitors are very good at detecting the latter.
4. The next step is more work than it looks
Every field, every decision, and every moment of ambiguity costs you conversions. Look hard at the gap between "I'm interested" and "I'm done":
- How many fields does the form have, and can you justify each one? Every field you can't justify should be deleted — you can ask later, once someone has a reason to answer.
- Is there exactly one primary action? Competing calls-to-action of equal weight force a decision the visitor didn't come to make, and the common outcome is that they make neither.
- Does the button say what happens next? "Get the free template" beats "Submit" because it tells someone what they're about to receive.
- What happens on mobile? Check on a real phone, not a resized browser window. Fixed-width elements, tiny tap targets, and forms that push the button off-screen are extremely common and nearly invisible from a desktop.
5. You don't have enough data to conclude anything yet
This is the most common cause and the least discussed, because it's the least satisfying.
At low traffic, conversion rate is dominated by noise. A page that has had a few dozen visitors and no signups has not told you it's broken — it has told you almost nothing at all. Founders routinely rewrite a page every few days on this kind of evidence, which destroys the ability to learn anything: each rewrite resets the clock, so the page never accumulates enough data to be judged.
What to do instead when traffic is thin: stop A/B testing, which needs far more traffic than you have. Change one substantial thing at a time, give it a fixed window, and lean on qualitative signal — session recordings, a one-question exit survey, or five conversations with people in your target audience. Five real conversations will teach you more than a split test that will never reach significance.
How to work through this
Do it in order, and stop at the first cause that clearly applies:
- Segment your traffic by source. If quality traffic converts, fix the traffic.
- Run the five-second test on three people who don't know your product.
- Find your proof and move it above the fold.
- Remove one thing from the path — a form field, a competing button, a step.
- Then wait. Give the change a fixed window before you judge it.
The reason for the order is that the causes aren't independent. Fixing friction on a page that nobody understands does nothing; rewriting a headline for visitors who were never going to buy does nothing. Each step rules out a class of explanation so that the next change you make is actually informative.
The part most pages get wrong
A landing page isn't finished at publish. It's a hypothesis about what a stranger needs to hear, and the only thing that can settle it is what real visitors do next.
That's the difference between treating a page as a deliverable and treating it as something you keep tuning: the first is finished when it looks right, the second is finished when it works. Most pages never make that transition — they get published, and then nobody ever looks at the data again.