How to write your first homepage hero

A stranger lands on your homepage. In about five seconds they decide whether to keep reading or close the tab, and most of that decision happens in the hero, the part above the fold with your headline and a line under it.

You can write a good one yourself. The work is not mysterious. What makes it hard is that a handful of small decisions along the way look obvious and are wrong, and you only notice the cost after you have shipped. So this is the whole process, start to finish, with those decision points marked. By the end you will be able to write your own hero and, more usefully, catch yourself before the easy wrong turn.

I will use one real site as a running example. It is a seed-stage B2B product. I have stripped the name. The mistakes are common enough that the name does not matter.

Key takeaways

• The hero has one job: orient a stranger in about five seconds. It is not a mission statement.
• Pick a shelf (the category they already know) and name one buyer. A line written to fit everyone reads as empty.
• Write a claim a competitor could not copy. Outcome-only and feature-only lines both fail.
• Test by covering your logo. If the hero could belong to anyone on your shelf, it is not specific enough yet.

1. Decide what the hero is for

The hero is not your mission statement. It is not where you say what you believe about your category or where the industry is heading. It is one thing: orientation for a stranger. In five seconds they should get what this is, whether it is for them, and a reason to scroll.

Here is the tell that you have drifted. Read your hero and ask whether it would sit just as comfortably on a competitor's site. If it would, it is describing a category, not you, and it is doing the wrong job. A first-timer writes the hero they would be proud to read. The hero that works is the one a stranger can use.

2. Pick the shelf

People understand a new thing by putting it on a shelf next to things they already know. Your hero has to tell them which shelf. If they cannot place you, they cannot compare you, and if they cannot compare you, they cannot decide you are worth anything. Choosing that shelf is a positioning decision, not a copywriting one, which is why a hero that fights you usually has a positioning problem underneath it.

Two wrong turns here, and they pull in opposite directions.

The first is assuming everyone already knows your category, so you write a broad, confident line that says nothing specific. This usually comes from copying a company much further along. A household name can run a vague hero because the brand does the orienting for them. You do not have that yet, so the same line that looks elegant on their site reads as empty on yours.

The second is the opposite move: inventing a new category name to sound different. It feels like differentiation. What it actually does is put you on a shelf nobody is shopping, so now you have to teach the shelf before you can sell the product.

The running example fails the first way. Its headline is the company name followed by "makes sense," and the line under it promises an experience that is "frictionless and continuous." Read it cold and you cannot tell what the product is or what it sits next to. The vagueness is not a copywriting problem. It is the symptom of a shelf that was never chosen.

3. Choose the one person it is for

Your product probably serves a few kinds of buyer. The hero cannot. A hero written to avoid alienating anyone speaks to no one, because no single reader sees themselves in it.

The fork is the fear of narrowing. It feels safer to say "for modern teams" than to name one role and one situation. It is not safer. The broad line is the one that gets skipped. The specific line is the one a reader stops on because it is plainly about them. Pick the single most important buyer for the hero. The others get their own space deeper in the site.

4. Write the claim

Now the headline itself. The most common failure has two faces, and both are easy to fall into.

The first is the outcome-only line: "scale without adding headcount," "close deals faster," "do more with less." It sounds like a benefit, but fifty companies could put their logo over it and nothing would change. It is not a claim about you. It is a claim about wanting good things.

The second is the feature-only line, a description of a mechanism with no reason to care, which lands as noise for anyone who does not already know the category.

A claim that works carries the mechanism and the reason together: the specific thing you do, and why that produces the outcome. The test is blunt. Read your headline and ask whether a competitor could say the exact same sentence. If yes, you have not written a claim yet. This is also where most AI-written copy lives, by the way. It is fluent, it is plausible, and it is interchangeable, because a model optimizing for "sounds like a homepage" produces the average of every homepage. The average is exactly what you are trying not to be.

5. Write the subhead

The line under the headline is not a second headline. It is the other half of the job. If the headline made the claim, the subhead says what the thing actually is and who it is for. If the headline named the category and audience, the subhead makes the claim. Between the two, a stranger should have all three: what it is, who it is for, why care.

The fork is restating the headline in new words because it feels like reinforcement. It is not. It is a wasted line. Cover the headline with your hand and read only the subhead. If it adds nothing the headline did not already say, rewrite it.

6. Earn the supporting elements

Around the headline you will want to add proof: a row of logos, a quote, a testimonial, a button. Each of these earns its place only if it does its job. At seed stage you often have not earned one yet, and an empty slot is better than a slot that pretends.

The running example breaks this twice.

Across the top it has a strip of recognizable logos. The natural read is "these are their customers." They are not. They are integrations, tools the product connects to. The strip borrows the credibility of a customer list, and the moment a visitor looks closer, or just thinks about it for a second, the borrow collapses into a small letdown at the top of the page. Worse, it quietly says the real thing out loud: we do not have customer logos worth showing, so we dressed up something that looks like them.

Just under that sits a large quote from a well-known analyst firm. Big, italic, prime real estate. The quote is not about this company. It is a generic statement about where the industry is going. They borrowed the logo of an authority without the thing the logo is supposed to signal, which is that the authority looked at them and had an opinion.

The pattern under both is the same: a slot filled with something that resembles proof rather than something that is proof. Borrowed trust that breaks on inspection costs you more than the blank space would have.

One more on this row. If you have stacked three calls to action, book a demo and start free and contact sales, you have not given the visitor a next step, you have given them a decision they did not come to make. Pick the one step you most want and let the rest go.

7. Test it yourself, properly

You are the worst possible judge of your own clarity, because you already know what the product does. So do not test for whether you like it. Test for whether a stranger gets it. Three that work:

Cover your logo and read the hero. Does it still read as yours, or could it belong to any company on your shelf? If it floats free of you, it is not specific enough.

Show it to someone outside your space for five seconds, then take it away and ask them what the company does. Their answer is your real hero, whatever you intended.

Read it out loud. The fluff and the borrowed phrases are easier to hear than to see. If a sentence sounds like it could open any homepage, it is opening yours badly.

The thread running through all of it

Look back at the wrong turns. The shelf that was never chosen. The buyer who was never named. The claim a competitor could have made. The logos that borrow trust they have not earned. They are all the same mistake wearing different clothes: an element placed on the page without anyone stopping to ask what it is for. The hero is small enough that you can ask that question of every single piece of it. Most founders never do, which is why most heroes say so little.

So here is the one to run right now. Cover your logo and read your own hero. Does it still sound like you, or could it be anyone on your shelf? Whatever you find, that is the thing to fix first.

That question, what is this for, is the whole job, and it does not stop at the hero. Every element on a site either earns its place or borrows trust it has not earned. Getting that right across the whole site, then holding it as you add to it over the next year, is the harder version of the same work. The hero is just the place it is easiest to see, and the best place to practice.

FAQ

What should a homepage hero include? What the product is, who it is for, and why they should care, plus one clear next step. That is it. The parts most heroes get wrong are not copywriting, they are the judgment calls underneath: which shelf, which buyer, which single claim.

How long should a homepage headline be? Short enough to read in one beat. Length is not the real metric. The real test is whether a competitor could say the exact same sentence. A short generic line fails it just as badly as a long one.

What makes a SaaS hero generic instead of specific? A claim that names an outcome anyone could promise ("do more with less") with no mechanism behind it, or a line copied from a company so well known the brand does the work the words should. The fix is a claim that carries both the thing you do and why it produces the result. Run the cover-the-logo test.

Should a hero have one call to action or several? One. Stacking book a demo, start free, and contact sales does not give the visitor options, it gives them a decision they did not come to make. Pick the single step you most want and let the rest live deeper in the site.