Your headline says what it is, not what it does
Cover your logo and read your headline. If a competitor could use it unchanged, it is not doing any work. The logo-swap test, and what to write instead.
Cover your logo and read your headline. If it could sit on a competitor's site without anyone noticing, it is not doing any work.
This is the most common finding in any audit and the cheapest to fix. Your headline names the category you are in. It does not say what changes for the reader. If your website headline is not converting, start here, because almost nothing else on the page gets read as reliably.
What does your headline say you do?
Run the logo-swap test. Screenshot your hero, cover the logo and the product name, then show it to someone outside your company. If they cannot tell you what you sell and who it is for, your headline is describing a category rather than an outcome.
- Cover the logo. Hide the wordmark, the product name, and any screenshot of the product.
- Read it aloud. If you stumble, so will a stranger.
- Ask two questions. What does this company do, and who is it for? Both answers must come from the headline alone.
- Swap in a competitor. Paste your headline under their logo. If it still reads fine, it belongs to neither of you.
Want this checked against the rest of your page? Revslip reads your hero the way a stranger does.
Why does a vague headline cost sales?
Because the headline is one of the few things on the page that reliably gets read. Nielsen Norman Group's analysis of real browsing behaviour found people take in roughly a fifth of the words on an average page. Your headline is most of that fifth.
Here is the arithmetic. From 45,237 real page views, Jakob Nielsen fitted a formula: about 25 seconds of fixed time on a page, plus 4.4 seconds per extra 100 words. At 250 words a minute, a visitor reads roughly 18% of anything you add.
Which gives you a ceiling. On the average page in that dataset, 593 words, a visitor has time for 28% of them, and only if they do nothing but read. Realistically nearer 20%. Readers get through half a page only at 111 words or fewer.
The question is not whether your headline is good writing. It is whether the twelve words a stranger definitely reads say what changes for them.
Why is my website headline not converting?
Usually because it describes what the product is rather than what it does for the reader. Nielsen Norman Group calls this the difference between the what and the how, and flags relative words like easier and more powerful as the sign that no value proposition has been decided.
Pavel Samsonov, writing for Nielsen Norman Group in July 2025, gives the test in one line: a strong value proposition names a goal the reader already wants, plus the thing that gets them there. His weak examples sit on half the sites we see. Ten exciting new features. More powerful than ever before.
Steve Krug made the same argument two decades earlier in Don't Make Me Think. A page has to answer "What is this?" before it earns the right to answer anything else. The words just got vaguer.
- 20% of the words on an average page actually get read
- 111 words the page length at which visitors read half of it
- 17% of page views in the dataset lasted under four seconds
Marking our own homework Those reading figures come from 25 instrumented browsers, data collected in 2005, and Nielsen notes himself that the participants were more literate than average. We use them because nothing better has replaced them, not because they settle it.
Should you put AI in your headline?
Probably not. In experimental surveys with more than 1,000 US adults across eight product and service categories, adding the words artificial intelligence to an otherwise identical description consistently lowered purchase intention. The mechanism the researchers identified was a drop in emotional trust.
The study is led by Mesut Cicek at Washington State University, with Dogan Gursoy and Lu Lu. In one experiment two groups saw identical smart television descriptions, differing only by that phrase. The group that saw it was less likely to buy.
The effect got stronger for what the authors call high-risk purchases: expensive electronics, medical devices, financial services. Cicek's own advice is blunt. Describe the features or the benefits, and avoid the AI buzzwords.
Awkward, because that is the phrase most likely added to your hero in the last two years.
How to rewrite it in an afternoon
Take the outcome your best customer describes when they explain why they stayed, and put that in the headline. Use the subhead for who it is for and how it works. Keep the whole hero under 30 words, so a stranger can finish reading it.
- Steal the customer's sentence. Take the exact phrasing from a support ticket, a review or a sales call. Not your tidied-up version.
- Name the change, not the category. Not "conversion optimisation platform". What is different on Friday that was not true on Monday.
- Cut the relative words. Easier, faster, smarter, more powerful. Each is a promise the reader has to finish for you.
- Let the subhead qualify. Who it is for, what it costs in effort, how long it takes.
- Run the logo-swap test again. If a competitor could still use it, go back to step one.
Not sure what your best customers actually say? Start with a free audit.
Telling a real lift from noise
Watch scroll depth and clicks on the primary button rather than overall conversion rate. A headline change moves the top of the page, and everything below it dilutes the effect. Give it two weeks and compare like periods, not last week against this one.
Most sites cannot A/B test a headline honestly. The traffic is not there, and a headline produces a small true effect and a large apparent one over a short run. If that is you, the low-traffic decision method is the honest alternative, and a five second test tells you more in an afternoon than an underpowered experiment will.
Across the 134 sites Revslip has audited, a hero stating a feature instead of an outcome is the most common finding in the copywriting category. Stated bias: sites come to us because someone already suspected a problem, so that is the frequency among worried owners.
When the headline is not your problem
If people read the hero, scroll, and leave further down, the headline did its job and something else is losing them. Same if your traffic arrives from the wrong search or the wrong ad. No headline rescues a visitor who was never going to buy.
The tell is where they go. A headline problem shows up as people leaving from the first screen. The shape of the leaving curve tells you which one you have.
Questions people ask about website headlines
Three come up repeatedly: how long a headline should be, whether to name the customer in it, and whether a brand name alone is enough. The answers follow from the reading data rather than style preference, which is why they are narrower than most advice on this.
Does headline length matter?
Less than clarity does. The reading research caps total words, not the headline specifically. A twelve-word headline a stranger understands beats a six-word one they have to decode.
Should the headline name the customer?
Only if naming them changes who reads on. "For agencies" earns its place when the product is wrong for everyone else. Otherwise it spends words qualifying people who were staying anyway.
Is a brand name alone ever enough?
Yes, once the brand does the explaining. The value-proposition argument bites hardest for products nobody has heard of, which is most of them. Until you are the exception, the headline carries it.
Wondering where this sits against everything else on your site? The conversion leak index puts the usual leaks in cost order.