What should a startup's landing page say?

Your landing page should answer three questions instantly: what you do, who it's for, and why it matters more than any alternative. Everything else—the team page, the pricing details, the blog—is secondary to those three answers. Get those right and the rest of the page is just reinforcement.

Start with the problem, not the product

Most early landing pages open with a product description. That's the wrong starting point. Users don't care what your product is—they care whether it solves a problem they already feel. The headline should name the pain or the outcome, not the mechanism. 'Stop losing sales leads in a spreadsheet' is stronger than 'CRM for small teams' because it speaks to the frustration the user already has, not the category you've decided you're in.

This isn't just copywriting advice—it's a test of whether you actually understand your users. Paul Graham's core point about startups is that you have to make something people want, and your landing page is the first place that claim gets stress-tested. If you can't say what problem you're solving in one punchy sentence, you probably haven't talked to enough users yet. The landing page forces clarity.

A useful exercise: write down the last three complaints your best users made before they found you. Your headline is probably hiding somewhere in those complaints. The landing page isn't marketing—it's a mirror of how well you understand the problem you're solving.

The hero section is doing almost all the work

Most visitors never scroll. That's not pessimism—it's a design constraint you have to work with. Your hero section (headline, subhead, and primary call to action) has to carry the entire argument. If a stranger can read just those three elements and understand what you do, who it's for, and what to do next, you've done your job. If they have to scroll to figure out the basics, you've already lost most of them.

The headline does one thing: name the outcome or the problem. The subhead does the next thing: add the 'for whom' and 'how.' The CTA should be a specific action, not a generic one. 'Start free' is weaker than 'Build your first pipeline free.' Specificity signals confidence and reduces the cognitive cost of clicking.

Avoid hero sections that feature a product screenshot behind a headline that says something like 'The future of work is here.' That combination communicates nothing. A real user reading that learns nothing actionable. Name the thing you do, for the person you do it for, and tell them exactly what happens when they click.

Social proof has to be specific to be credible

Generic testimonials ('This product is amazing!' — Jane D.) are worse than no testimonials at all because they signal that you couldn't get a real quote. The social proof on your landing page should be specific enough that a skeptical visitor could verify the claim or at least imagine the real person behind it. 'We reduced churn by 18% in the first 60 days' from a named company in a recognizable industry is worth a dozen generic superlatives.

If you're early and don't have impressive logos yet, use specificity to compensate. A quote from one user that describes their exact situation—what they tried before, what changed, what the measurable result was—is more persuasive than six vague endorsements. Founder-stage landing pages often skip social proof entirely because they don't feel like they have anything impressive enough. Ship whatever you have. A specific quote from one real user beats a blank section every time.

For developer tools especially, the most powerful social proof is often a real usage example. Showing what a user actually built with your tool—with their name attached—does more work than a star rating. The goal is to make the next potential user think 'that's my situation exactly.'

Design signals whether you're serious

Visitors make a trust judgment about your company in milliseconds based on visual design, and that judgment affects whether they read anything else on the page. This doesn't mean you need a sophisticated design system—it means you need a coherent one. Consistent typography, a restrained color palette, and intentional whitespace communicate that you're a serious operator who pays attention to details. Inconsistency communicates the opposite.

Garry Tan's design work on gstack shows what this looks like in practice for a developer tool: a single clear aesthetic direction (industrial/utilitarian in that case), two or three typefaces with defined roles, and decoration that serves function rather than filling space. The aesthetic doesn't have to be beautiful—it has to be intentional. Founders in early-stage companies often underestimate how quickly a visitor decides whether to trust them, and how much of that decision is visual before it's rational.

Practically: pick one font pairing and use it consistently. Pick a primary color and one accent. Remove any element you can't justify with 'this helps the visitor understand or take action.' Clutter doesn't signal richness—it signals distraction. The companies whose landing pages work best tend to look like they removed things until only the essential remained.

The call to action should match where users are in their decision

A single landing page often serves visitors at very different stages: someone who heard about you from a friend and is ready to sign up, and someone who just found you from a search and is still figuring out what you do. Your primary CTA serves the first group; you need a secondary path for the second. 'Get started free' plus 'See how it works' is a common and effective pairing. The first captures ready buyers; the second keeps curious browsers from leaving with nothing.

Be precise about what happens after the click. 'Get started' is vague. 'Create your free account—no credit card' is specific and removes the most common friction point in a single phrase. If there's a free trial, say the length. If setup takes under five minutes, say that too. Every unspoken question a visitor has is a reason not to click, and you can preempt most of them with a sentence or two near the CTA.

One thing to avoid: making your primary CTA 'Book a demo' when your product can sell itself with a free trial. Demo-first flows introduce friction and delay, and they work best for enterprise products with long sales cycles and complex implementation. If a user can experience value in your product within minutes, let them. The landing page's job is to get them there as fast as possible.

“There is no boss to trick, only users, and all users care about is whether your product does what they want.”

— Paul Graham, source

The one thing to do

Rewrite your headline today so it names the specific outcome or problem your best user cares about—then read it aloud to someone who's never heard of your product and see if they immediately understand what you do.

Frequently asked questions

How long should a startup landing page be?

Long enough to answer every objection a skeptical visitor has, short enough that nothing is redundant. For most early-stage products, that's a hero section, a brief explanation of how it works, social proof, and a final CTA—often under 800 words of copy. If you're still figuring out your positioning, start shorter and add sections only when users tell you they're confused about something specific.

Should I show pricing on my landing page?

Yes, if you have a clear pricing model and your price is a strength or a differentiator. No, if you're still testing pricing or if the right price depends heavily on the user's situation. Hiding pricing entirely can signal that the product is expensive or complicated; showing it too early in a confusing way creates its own friction. When in doubt, at least indicate whether there's a free tier or trial.

What if my product is hard to explain in one sentence?

That's a product positioning problem, not a copywriting problem. If you genuinely cannot describe what you do in a sentence, you likely haven't narrowed your use case enough yet. Talk to your best three users and ask them to explain what you do to a colleague—the language they use is almost always clearer than what's on your current landing page.

How do I know if my landing page is working?

The primary metric is conversion rate from unique visitors to your primary CTA action—signup, trial start, or waitlist join. Anything below 2-3% on targeted traffic suggests a messaging problem. Watch where people stop scrolling and what they click using a heatmap tool. Run your headline past five people who fit your target user profile and ask them to repeat back what the product does—their answers tell you more than any A/B test.

Sources

More playbook answers · Growth Prophet home