Seven homepage mistakes that quietly drop your SaaS conversion rate to half of what it could be.

The hook

Unbounce's 2024 Conversion Benchmark Report tracked 41,000 landing pages, 464M visitors, and 57M conversions. SaaS came in at the bottom of every tracked vertical with a 3.8% median. Financial services were 8.4%. Events were 12.3%. SaaS, the most measurable industry in tech, is also the worst converting.

Why? Mostly because seven specific mistakes show up on roughly 80% of SaaS homepages. Fix any two and you'll likely beat your category median. Fix five and you're in unicorn territory.

Mistake 1: A headline that could belong to anyone

The most common mistake by a mile. The hero reads like it was generated by a vague-business-words API.

Real examples I've collected from SaaS homepages this year:

• "Smarter solutions for modern teams."
• "Reimagining how businesses work."
• "The intelligent platform unifying enterprise data."
• "Where innovation meets execution."

Each of those could be enterprise software, a consulting firm, a CRM, or a fitness app. They tell the visitor nothing. CXL's research found 53% of audited SaaS pages led with feature-first heroes rather than outcomes. April Dunford's positioning frame: customers buy outcomes, not capabilities.

The fix. Use one of these formulas:

• "[Outcome verb] for [specific audience]", Loom: "One video is worth a thousand words"
• "The [category] for [unique mechanism]", Linear: "The product development system for teams and agents"
• "Do [JTBD], not [old time-consuming task]", Fathom: "A Google Analytics alternative without compromise"

The 5-second test: show your headline to someone outside your industry. If they can repeat back what you do, ship it. If they can't, rewrite it.

Mistake 2: A hero without the product visible

Stock illustrations of diverse smiling people at laptops are the visual equivalent of "innovative solutions." They don't sell software. They signal "we couldn't think of what to put here."

Fix. Put the product in the hero. Three options that all work:

1. A real screenshot of your product doing its main thing. Not a sanitized "looks-like-a-product-but-isn't" mockup. Linear shows actual Linear issues.
2. An autoplaying muted video of the product in use. Loom does this. The CTA is implicit, watch the demo by being on the page.
3. An interactive demo embedded in the page. Notion has 3D-rendered integration icons floating around a working interface. Vercel renders live AI model usage telemetry directly into the hero.

What never works: an abstract gradient with floating UI elements that don't correspond to any real screen. Visitors can tell.

Mistake 3: Multiple competing CTAs above the fold

Open a low-converting SaaS homepage and count the CTAs in the first viewport. Usually three to five: "Get a demo," "Start free trial," "Watch video," "Talk to sales," "Learn more."

Each one fights the others. Decision fatigue triggers. The visitor picks the easiest option (closing the tab).

CXL's research is unambiguous: multiple CTAs reduce action. One primary, optionally one secondary, never more.

Fix. Pick the single action you want most. For most B2B SaaS, that's either "Start free trial" or "Book a demo." Everything else gets de-emphasized. Stripe: "Sign up" + "Sign up with Google" (same action, two methods). Linear: "Start building" + small "Pricing" link. Vercel: "Start Deploying" + "Get a Demo." Two CTAs maximum, one clearly primary.

The one-CTA test

If you can't say in 5 seconds which CTA on your homepage is the primary one, neither can your visitor. Visual hierarchy: filled button = primary, outline = secondary, text link = tertiary. Don't make two CTAs the same weight.

Mistake 4: Copy written at college reading level

Unbounce's 2024 data is brutal on this one: pages at 5th–7th grade reading level converted at 11.1%. Pages at professional reading level converted at 5.3%. That's 2x. The correlation has strengthened 62% since 2020.

What "professional reading level" looks like:

"Our enterprise-grade platform leverages machine learning to optimize workflow efficiency across your organization's distributed teams, enabling seamless collaboration and accelerated time-to-value."

What 5th-grade copy looks like:

"Run your team's projects in one place. Faster than email, simpler than Slack."

Most SaaS marketers default to the first because it sounds professional. It actually signals you don't know what your product does well enough to explain it simply. Difficult words alone correlated with a 24.3% conversion drop in the same study.

Fix. Run your homepage copy through Hemingway. Aim for grade 6 or below. Strip every adjective. Remove every "leverage," "robust," "seamless," "innovative," "platform." Most of those words can be deleted with no information loss.

Mistake 5: Hidden pricing

62% of B2B buyers disqualify vendors who hide pricing before engaging sales (TrustRadius 2025). The pricing page is the second-most-visited page on 80% of SaaS sites (Profitwell).

Hidden pricing means "Contact sales for a quote" with no number. The argument for it, "we lose pricing power if competitors see it", is real but small. The cost of disqualifying 62% of inbound is enormous.

Fix. Three rules:

1. Pricing in the nav. Stripe, Linear, Notion, Vercel, Loom all have a top-nav "Pricing" link.
2. A starting-at anchor even for custom pricing. "Starting at $499/mo" + "Contact sales for enterprise" outperforms pure "Contact sales."
3. Three tiers, one highlighted as recommended. Three-tier pages convert ~1.4x two-tier. Pages without a recommended tier convert 22% worse.

Outcome-led pricing pages convert 34% better than feature-led (HubSpot 2024). "Best for marketing teams that..." beats "Includes 50 users and 20GB storage."

Mistake 6: Generic or buried social proof

Three patterns that kill social proof:

The "Trusted by industry leaders" graveyard. No actual logos. Just the phrase. Visitors read this as "we have no real customers."

Logos of companies nobody recognizes. A logo bar with eight obscure regional firms reads worse than no logo bar. Cut to the three best.

Burying testimonials in the footer. CXL's eye-tracking research is clear: testimonials with photos held attention longest (1.06s avg fixation) when placed near a CTA or at a point of friction. In the footer, they're invisible.

Fix. Three rules:

1. Logo bar within the first scroll. Stripe's scrolling marquee, Loom's static wall, Vercel's inline single-sentence quote with logo.
2. Named testimonials with photo, role, and company. "Sarah Chen, Head of Engineering, Linear" beats "S.C., Tech Company."
3. Specific numbers where possible. Notion: "Over 100M users worldwide • 62% of Fortune 100 • Over 50% of YC companies." Specificity beats vague claims.

Peep Laja: "Push back when your designer recommends hiding your social proof at the bottom every single time. Use social proof as supporting copy near a CTA or at a point of friction."

Mistake 7: A form that asks for everything

Unbounce 2026 data: 3-field forms convert at 10.1%. 9-field forms convert at 3.6%. That's 2.8x. Every field has a measurable conversion cost.

The typical SaaS sign-up form asks for: name, email, company, role, company size, phone, country, password. Eight fields. The visitor sees eight fields and thinks: "this is going to lead to a sales call." Half of them close the tab.

Fix. Strip to the bare minimum.

• Free trial signup: email + password. (You can ask for company info inside the product after they're in.)
• Demo request: name + email + company + one optional context field.
• Newsletter: email only.

For everything else, ask later. You can collect company size, role, and use case inside the onboarding flow, and the visitor is already committed by then. Asking everything upfront is a conversion tax.

The form math

A SaaS site with 10,000 monthly visitors and an 8-field form converting at 3.6% gets 360 signups/month. Same traffic, 3-field form converting at 10.1%, gets 1,010 signups. That's 650 more qualified people in your funnel every month from one design change.

Bonus: the meta-mistake most pages make

The seven above are mechanical. The meta-mistake is structural: the page is trying to do too many jobs.

A SaaS homepage tries to convert investors, qualify enterprise buyers, sign up self-serve users, hire engineers, explain the product to journalists, justify the pricing to procurement, and tell the founder's story. Each of those visitors needs different copy, different proof, different CTAs.

The high-converting homepages pick one visitor and serve them ruthlessly. Linear is for product engineers. Stripe is for any business with online revenue. Loom is for anyone explaining things asynchronously at work. Each has a clear ICP and writes for that person.

Fix. Pick your single best-fit visitor. Write the homepage for that person. Move everything else to dedicated pages, investor info on a /investors page, careers on /careers, press on /press. The homepage gets to be focused.

The 7-mistake audit

Open your own homepage now. Score yourself.

Mistake	Pass	Fail
1. Generic headline	Specific, outcome-led	Could belong to any company
2. No product in hero	Real screenshot or video	Stock illustration or abstract
3. Multiple competing CTAs	One primary, max one secondary	3+ CTAs in first viewport
4. College-level copy	Hemingway grade 6 or below	"Leverages," "robust," "seamless"
5. Hidden pricing	Pricing in nav with starting anchor	"Contact sales" with no number
6. Generic / buried social proof	Named logos + photo testimonials in first scroll	Generic phrases or footer-only
7. Too many form fields	3 or fewer for trials	6+ fields for any sign-up

Score 5 or higher (failing): your conversion rate is likely below the 3.8% SaaS median. Score 6+ passing: you're probably in the top quartile already.