The Ultimate SaaS Landing Page CRO Checklist
Stop guessing what's hurting your signup rate. Score your page against 79 critical CRO checks and walk away with a ranked list of exactly what to fix next.
Above the Fold & Value Proposition
Five seconds is the budget. Text your URL to three friends, ask what your product does, wait. The answers will hurt. Use them.
Abstract illustrations of people high-fiving in front of a laptop are an admission you don't have a screenshot worth showing. Buyers came to evaluate software, not to admire your art direction. If your product genuinely doesn't look impressive yet, that's a product problem dressed as a design problem. Don't paper over it with illustrations.
'The modern way to collaborate' is a sentence no one searched for. Plant a flag in an existing category — 'project management for engineering teams' — then add what makes you different from the other twenty options in it. The temptation to invent a new category is almost always a mistake. New categories take a decade to build, and you don't have a decade.
Every 'Learn more,' every nav link, every 'watch demo' is a small invitation to leave the conversion path. If a link can't outrank your primary CTA for attention, it doesn't belong in the first screen.
The hero image or video is almost always the largest contentful paint element. It's also the easiest CWV win on most pages. Fix this before you touch anything else on the speed checklist — the rest is small change.
The moment a visitor starts scrolling, they should see proof. Logos at the bottom of the page do nothing because most visitors never see them. If your logos aren't impressive yet, use industry-relevant ones over famous ones. Relevance converts better than fame.
Your team calls it 'AI-powered workflow orchestration.' Your buyers type 'project management software.' Cross-check your hero against your Search Console queries. The gap is wider than you think.
A real number from a real customer beats every adjective in your thesaurus. If you can't source a defensible number, don't fake one — a missing number is better than one that won't survive a sales call.
Headline & Messaging Clarity
The cheapest CRO test that exists and almost nobody runs it. Five strangers, five seconds, 'what does this do?' If three of five can't answer, the headline is too clever. 'Clever' is the most expensive word in landing page copy. It's what founders reach for when they're bored of their own product. Boring and clear beats clever and ambiguous every time.
'Close deals faster' beats 'AI-powered CRM with pipeline analytics.' The mechanism belongs in section three, where prospects who already care about the outcome go looking for proof. Exception: in developer tools and infra, the mechanism is the outcome. Engineers don't trust pages that hide it.
The most common pattern we see: a homepage selling 'save time' to operators AND 'grow revenue' to executives AND 'ship faster' to engineers. The result converts none of them. Split into separate landing pages by buyer. Route paid traffic to the matching page. The lift from message-match alone usually beats every other change combined.
'Built for B2B SaaS founders.' 'Made for product marketers at Series A startups.' The right visitor feels seen. The wrong one self-disqualifies. The second is undervalued — wrong-fit visitors pollute your funnel metrics and make it harder to learn what's working.
'The Notion alternative for engineering teams' captures someone the moment they land. The visitor has already done half the buying work — they know the category, they know which incumbent they don't like, they're looking for permission to leave.
Every 'users can' becomes 'you can.' Read your page out loud. The moment you hear yourself describing a product instead of talking to a person, mark it and rewrite.
'Reporting' is a label. 'See where every dollar of your ad spend went' is a section header. Most visitors only read your H2s. If a skimmer reading only your H2s couldn't reconstruct the offer, your headers are doing decoration work instead of selling work.
'Best.' 'Leading.' 'World-class.' 'Next-generation.' The buyer's brain skips every one. Replace each with a fact, a quote, or a number — or delete it. 'Trusted by 8,000 teams' beats 'the most trusted platform' because one is checkable and the other is wallpaper.
Even technical buyers skim. The best B2B SaaS pages read like a smart friend explaining the product over coffee. Run Hemingway. Cut anything above grade 9.
Social Proof & Trust
'— Marketing Director' reads as fabricated, because half the time it is. One fully attributed testimonial is worth ten anonymous ones. If you have only one, lead with it. Don't dilute it.
'Saved us 12 hours a week' beats 'great product, highly recommend' by an enormous margin. Make it a quarterly habit: ask your three happiest customers for one number they'd put their name to. You'll get at least one.
A visitor from the Shopify ecosystem is moved by Shopify-ecosystem logos. They're unmoved by your one Fortune 500 client from a totally different vertical. Relevance > fame. Segment the logo strip by referral source if you can — the engineering work is small, the lift is real.
A G2 badge that doesn't click through looks Photoshopped. A G2 badge linking to a page where your last review is from 2022 is worse than no badge. Check each one. Sounds trivial. Isn't.
If you serve businesses, name what you've got — even just a 'privacy policy' link or a 'data hosted in EU' line near pricing helps. SOC2 and ISO 27001 only matter once you sell to enterprise. Until then, show the smaller wins you do have rather than hiding behind silence — buyers assume the absence is the answer.
If you have any happy customer, ask for a 30-second clip recorded on their phone. One scrappy testimonial outperforms zero polished ones. Tone and body language are harder to fake than text, and at indie scale you don't need flagship logos — you need proof a real person uses this.
If you have one customer with a real outcome, that's enough for one case study at this stage. Problem, solution, result with a number in the headline — and don't bury it three clicks deep. One detailed story outperforms five generic testimonials. Don't wait until you have ten.
Product Demonstration
Marketing renders look like aspirations. Annotated screenshots — with arrows, callouts, captions explaining what each piece does — look like proof. Buyers want proof.
Hero or section two. Not button-gated. Not 'scroll down to find it.' The demo at the bottom of your page is a demo that doesn't exist.
Wistia's data is consistent: most viewers drop after 60 seconds. The best demos hook in five seconds and deliver the core 'aha' by 45. If yours is three minutes, cut it in half. If it's six, you're treating it as documentation, not as a sales tool.
Three steps fit in working memory. More steps signal complexity, even when there isn't any. If you genuinely can't compress to three, the product is more complicated than your buyers want — and that's the real problem.
The single most common screenshot mistake we see. 'Project Alpha' and 'user@example.com' make the product feel like a demo for a demo. Realistic project names, plausible numbers, real-looking dates make it feel like a working product. Use your own dogfood account if you have to.
For any product where signup is a heavy commitment, this is often the single biggest unlock in the entire checklist. Storylane, Arcade, Navattic — all one-day builds. If you've never seriously considered adding one, you're probably leaving conversions on the table.
'Tired of building reports manually every Monday?' leads. 'Automated reporting dashboard' follows. Problem-first framing makes the feature feel inevitable. Feature-first leaves the reader asking 'do I need this?' — which is the wrong question to leave hanging.
'Works with Slack, HubSpot, Stripe' removes a major buying objection. Text lists are skipped. Logos are absorbed. If your top integration is also your biggest competitor, lead with it anyway — buyers care more about fit with their stack than your positioning concerns.
Pricing Page & Plan Presentation
'Contact us for pricing' works for two kinds of companies: pure enterprise, and brands strong enough that buyers tolerate the friction. If you're neither, 'Contact us' is just SMB buyers reading 'we'll figure out how much you can pay' and leaving. Even ranges work. The fear that publishing prices commoditizes you is almost always smaller than the cost of disqualifying yourself.
More than three columns triggers analysis paralysis. Hide the unusual plans behind a toggle. The pricing page exists to make a choice easy, not to display your full SKU catalog. Founders resist this because they think more plans = more revenue. The opposite is usually true.
A 'Most popular' badge, a contrasting background, a thicker border. Decision fatigue is real, and most buyers want to be told which plan to pick. Be the guide. Don't be coy about which plan you want them on.
Long lists train the eye to skim and make plans feel interchangeable. Show only what's different between plans. The full matrix lives below in a comparison table for the small minority who'll actually read it.
'Save $X/year' beats '-20%' because the dollar feels real. If invited viewers are free, say so. If everyone is billed, make that obvious. Pricing surprise turns a happy trial into an angry churn faster than anything else.
The enterprise plan with only 'Contact sales' loses the buyer who's a fit but isn't ready to talk. Add 'Or start a free trial' as a secondary action.
Reversibility shrinks decision risk. The buyers who would have abused it were never going to pay long-term anyway. The cost is tiny; the lift is real.
Pre-answer the five to eight questions buyers ask before paying: refund policy, what happens at the seat limit, what counts as a 'user,' are there setup fees. Pull from your actual support inbox. If you've answered the same question more than three times this month, it belongs here.
Even if 99% of your buyers are self-serve, a 'Contact sales' option anchors the lower plans as bargains and captures the occasional high-ACV deal that walks in. Don't omit it because you've never had one — you'll figure out the enterprise product when the first lead arrives.
Primary CTA & Signup Flow
'Start my free trial' outperforms 'Get started' and 'Sign up' because the user mentally rehearses the action. Tiny copy change, consistent lift across B2B SaaS pages.
If your button color matches your nav or header brand color, it visually disappears. The CTA should be the single highest-contrast element on the page — the one thing the eye lands on first.
Most converters click in the bottom half of the page, after they've read the proof. One CTA in the hero is not enough — repeat it after each meaningful section so users never have to scroll back.
Every additional field drops conversion (HubSpot and Unbounce both have data on this). The signup form is not the place to qualify leads. The signup form is the place to capture them. Qualify in onboarding.
SSO eliminates typing friction and lifts B2B signup completion meaningfully when it's there. But building Google/Microsoft sign-in is a Series A problem, not a micro-SaaS one. If you have it, make it the visually dominant signup option. If you don't, ship email + password well first.
Five-word edit. Consistent lift. If your trial does require a card, ask yourself why. The conversions you lose to that requirement usually outnumber the bad-fit users you screen out.
Enterprise buyers want demos. SMBs want to try it themselves. If your average revenue per customer is north of ~$200/mo, forcing one path leaves money on the table — offer both. Below that, self-serve only is fine and removes a support burden you don't need yet.
'Free for 14 days, then $X/mo. Cancel anytime.' The visitors most likely to convert are also the most cautious about hidden commitments. Tell them what's behind the door before they open it.
Lead Capture & Forms
For 'book a demo': email, name, company. Done. Enrich the rest from the email domain or ask on the call. The marketing team's desire to qualify at the form is almost always more expensive than the leads it filters out.
If you do demos at all, embed Calendly or Chili Piper. The 'we'll be in touch' flow is where demo leads die — the friction of waiting for an email, replying, scheduling is enough time for intent to evaporate. If you don't do demos yet, skip this.
Phone number on an e-book form is how you get a database full of '555-555-5555.' Save deeper qualification for actual buying-intent forms.
Capture intent at the moment of value, and again after the user finishes scrolling. Pair with a real reason to subscribe. 'Weekly CRO tips for SaaS founders' beats 'subscribe to our newsletter' — the first names what they get, the second names what you want.
Trust dies fast. If the promised PDF or trial link isn't in their inbox within a minute, they assume it's broken and bounce. Test monthly. These things break silently.
The user just signaled high intent. Capitalize on it. 'Check your inbox' or 'Book your onboarding call' or 'Watch this 5-minute video' — anything that keeps momentum beats a digital handshake and a closed door.
Objection Handling & FAQ
Pull from sales calls and support tickets, not from what's comfortable to answer. If your FAQ doesn't include at least one question that makes your team slightly uncomfortable to answer publicly, it isn't doing its job.
Buyers shortlist three to five vendors. If you don't position against the alternatives, they'll assume you're undifferentiated and pick on price — and you'll lose, because someone is always cheaper. Name the competitor. Acknowledge what they're good at. Explain when you're the better choice. Founders avoid this because it feels aggressive. The avoidance costs more than the awkwardness.
The number one reason buyers stay on inferior tools is switching cost. 'We'll import your data from [competitor] in under an hour' is one of the most powerful sentences you can put on a B2B SaaS page. If you do the migration for them, say so prominently.
If yours is fast, brag. If it's slow, justify it with what you do during onboarding. The vague answer is always worse than the honest one, even when the honest one is 'this takes time.'
One sentence on the page — 'export your data anytime in standard formats' — moves more deals than you'd expect. Even if the export UI isn't built yet, commit to it. Mature B2B buyers look for this, and the absence of the statement is itself a signal.
Usage- and seat-based pricing punishes buyers who can't model their year-two cost. One worked example ('a team of 10 at average usage = $X/mo') is table stakes for any growth-priced product. Without it, buyers assume the worst and bake the worst into their budget.
Page Speed & Core Web Vitals
Users bounce roughly 25% more per extra second (Google's published research). LCP is usually the hero. Compress aggressively. Lazy-load everything else. Check your numbers in Search Console, not just locally.
Reserve image dimensions. Avoid late-injected banners. Pre-load fonts. Often the easiest CWV to fix and the most overlooked, because developers don't notice the jumps on their fast local connections.
Sluggish buttons make your product feel slow before users even sign up. Strip heavy third-party scripts. Defer non-critical JS. Audit your tag manager — half the scripts in there are probably doing nothing useful.
A 200KB WebP versus a 1.2MB PNG cuts LCP by half a second on slow connections. Every modern image pipeline supports this. There is no excuse.
Synchronous third-party scripts block paint and inflate INP. Audit your tag manager — most things load fine async or after user interaction.
If you're on Vercel, Netlify, or Cloudflare Pages, edge delivery is the default — but custom domains, image hosts pointing at origin, or misconfigured rewrites can quietly bypass it. Check your asset URLs are hitting the CDN, not your origin. If you're not on a modern host yet, fix that first; rolling your own CDN is bigger work than it sounds.
FOIT (flash of invisible text) hurts perceived speed and CLS. font-display: swap and <link rel=preload> together eliminate the worst font-loading regressions.
Mobile Experience
Tiny mobile type makes users pinch-zoom or bounce. 16px body is the absolute floor — 18px is better for B2B audiences over 35 (which is most of them).
Thumb-reach matters. A full-width button is the easiest tap target and signals 'the next step' louder than a centered button surrounded by air. Both improve mobile conversion.
Don't hide signup three taps deep. Many users open the menu specifically looking for it. Keep the CTA visible in the mobile header too, if space allows.
Mobile data, battery, and tab freezing kill mobile autoplay. Poster image, play overlay, load video only on tap. Saves the visit, the conversion, and your CWV scores simultaneously.
Horizontal-scroll pricing on mobile is unusable. One vertical card per plan, recommended plan first. Check it on a real phone, not by resizing your browser window — half your traffic might be seeing a broken page right now.
Onboarding & Activation Hooks
'Check your email to verify' is where activations go to die. Every second the user spends switching apps is a second they might decide not to come back. Magic links, SSO, auto-login. Verify in the background.
Pick the single behavior most correlated with retention. Walk users there. Don't show them everything the product can do — show them the one thing that makes them say 'oh, I get it.' Everything else waits until session two.
The most common new-user paralysis happens when the email says 'do X first' and the tour says 'do Y first.' Two 'do this firsts' equals zero. Pick one path. Reinforce it from both channels.
Long tours train users to click 'Skip' forever — and on every product they ever use afterward, so you're damaging the entire SaaS ecosystem one bad tour at a time. The best tours are under 60 seconds.
An empty dashboard with 'Import sample data' beats an empty dashboard with a generic help link. Empty states are the first product surface most users see. Treat them as onboarding copy, not as design afterthoughts.
Letting users see the product populated removes the 'I have to do work before I see value' friction that kills activation. Often the difference between a 20% activation rate and a 60% one.