We Analyzed 50 SaaS Landing Pages — Here's What's Killing Their Conversions

Most SaaS companies spend thousands driving traffic to their landing pages. Very few bother checking if those pages actually convert.

We used Leak Detector to run AI-powered audits on 50 SaaS landing pages — from billion-dollar companies like Stripe and Notion to fast-growing indie tools like Lemon Squeezy and Cal.com. Each page was scored across 8 categories: headline clarity, CTA effectiveness, social proof, forms, visual hierarchy, trust signals, mobile experience, and performance.

The results were worse than we expected. Here's what the data says.

The Dataset

We selected 48 SaaS landing pages across 6 categories:

Batch	What We Tested	Pages
Established SaaS	Stripe, Notion, Figma, Slack, Vercel, Calendly, Airtable, Lemlist, Crisp	10
Landing Page Builders	Unbounce, Instapage, Leadpages, Carrd, Framer, Webflow, Typedream	7
Indie SaaS	Tally, Plausible, Buttondown, Cal.com, Beehiiv, Gumroad, Lemon Squeezy, PostHog	11
Marketing & CRO Tools	Hotjar, Mixpanel, Amplitude, Semrush, Ahrefs, Mailchimp, ConvertKit, ActiveCampaign, Intercom, Brevo	10
Pricing Pages	Notion, Linear, Vercel, Figma, Slack (pricing-specific pages)	5
French SaaS	Alan, Qonto, Pennylane, PayFit, Spendesk	5

Out of 48 pages, 40 returned full audit results. 8 either timed out (heavy JavaScript rendering) or blocked access — which is itself a finding we'll discuss.

The Big Picture: Average Score Is 73/100

The average score across all 40 successful audits was 72.7 out of 100. That sounds decent until you realize what it means: the typical SaaS landing page is leaving roughly a quarter of its potential conversions on the table.

The best score? Lemon Squeezy at 82/100 — a clean, focused page with strong social proof and a clear value proposition.

The worst? Slack's pricing page at 25/100 — an enterprise-focused page that fails on almost every front for individual visitors trying to evaluate the product.

Where SaaS Landing Pages Fail: Category Breakdown

Here's where things get interesting. While the overall scores cluster around 70-78, the category-level data reveals dramatic gaps:

Category	Avg Score	% Failing (<70)	Verdict
CTA	65.1	57.5%	Biggest conversion killer
Social Proof	60.1	59%	Most neglected element
Forms	64.9	47%	Nearly half have no capture path
Performance	68.5	40%	Speed still an afterthought
Headline	74.5	30%	Better, but not great
Visual Hierarchy	78.7	7.5%	Generally solid
Trust Signals	80.5	10%	SaaS companies get this right
Mobile	81.5	2.5%	Mostly responsive in 2026

Three categories stand out as systemic problems.

Finding #1: 57% of SaaS Pages Have Weak CTAs

This was the most surprising result. More than half of the pages we analyzed had CTAs that scored below 70 — including companies whose entire business depends on getting visitors to click a button.

The most common issues we flagged:

Generic button text with no value proposition. "Book a demo" and "Get started" appeared everywhere. These CTAs tell visitors what to do but not what they'll get. The pages that scored highest on CTA (85+) used benefit-driven copy: language that communicated the outcome, not just the action.

Competing CTAs creating decision paralysis. Spendesk, for example, shows both "Book a demo" and "Tour the Spendesk app" side by side. When visitors face two equal-weight choices, many choose neither. The strongest pages we audited had one dominant CTA with a clear visual hierarchy over any secondary options.

No urgency or differentiation. Almost none of the pages tested created any sense of time pressure or exclusivity. In a category where every competitor says "Get started free," the ones who add specific context — trial length, no credit card required, time to value — consistently scored higher.

CTA scores ranged from 15 to 90 across our sample, the widest spread of any category. This means it's both the area where pages fail the hardest and where improvements would yield the biggest gains.

Finding #2: 59% Have Inadequate Social Proof

Nearly 6 out of 10 pages scored below 70 on social proof — making it the category with the highest failure rate.

The pattern we noticed: most SaaS pages either have no social proof above the fold, or they rely exclusively on logo bars. Logo bars are fine, but they're table stakes in 2026. Visitors have learned to ignore a row of grayscale company logos.

What separated the high scorers from the low:

Specific numbers beat vague claims. Pages that displayed concrete usage data ("5,000+ finance teams" on Spendesk, or specific review scores like "4.7/5") consistently outscored those with just logos. Plausible.io and Lemon Squeezy both scored 85+ on social proof by leading with specific, verifiable metrics.

Testimonials with attribution are rare. Across all 40 pages, detailed customer testimonials — with real names, real companies, real results — were uncommon. Most pages either skipped testimonials entirely or used generic quotes without attribution. This is a missed opportunity: a single specific testimonial with a measurable result outperforms a wall of logos.

Some pages scored 0 on social proof. Zero. These were pages where no customer evidence was visible anywhere in the initial viewport. For a SaaS product asking for money or personal information, that's a hard sell.

Finding #3: 47% Have No Effective Lead Capture

"No lead generation form present" was the single most flagged issue in our entire study — appearing on nearly half of all pages analyzed.

Now, context matters here. Some of these pages (Stripe, Linear) run a product-led growth model where the homepage's job is to drive signups, not capture leads through forms. The "Book a demo" CTA is their form.

But for the majority — especially tools competing in crowded markets — having zero low-friction capture mechanism means they're only converting visitors who are ready to buy right now. Everyone else bounces and is gone forever.

The pages that scored highest on forms offered multiple entry points at different commitment levels: a free tool, a newsletter signup, a resource download, or a calculator — alongside the primary conversion CTA. This multi-tier approach captures visitors at different stages of the buying journey.

The Pricing Page Problem

The most striking finding in our batch analysis: pricing pages scored an average of 57/100 — seventeen points below the overall average of 74.

Batch	Average Score
Indie SaaS	75
Established SaaS	74
Landing Page Builders	74
French SaaS	74
Marketing/CRO Tools	73
Pricing Pages	57

Slack's pricing page came in at 25/100 — the lowest score in the entire study. The page presented a complex enterprise pricing structure with no visible social proof, no testimonials, and poor mobile optimization for what is often the highest-intent page on a SaaS website.

This matters because pricing pages typically see the most purchase-ready traffic. A visitor on your pricing page has already decided they're interested. If that page scores 57 while your homepage scores 74, you're losing conversions at exactly the wrong moment.

The Landing Page Builder Irony

Here's a finding we didn't expect: landing page builders scored exactly average (74/100).

Unbounce, Instapage, Leadpages, Framer — these companies literally sell the tools to build high-converting landing pages. You'd expect their own pages to be best-in-class. They're not. They're middle of the pack.

Instapage led the builder category at 78. Carrd and Typedream came in at 72. For companies whose core promise is "we help you convert better," performing at the industry average is an awkward look.

The main weakness across builder pages was, ironically, social proof. Several builder landing pages lacked specific conversion data or case studies — the exact kind of evidence their own customers need to see.

17% of Pages Couldn't Even Be Analyzed

Eight of our 48 target pages — including Linear, Slack, Webflow, and Pika — either timed out after 30 seconds or blocked automated access entirely.

These are all JavaScript-heavy single-page applications that take too long to fully render. While this doesn't directly measure conversion quality, it signals a performance problem. If an automated tool can't fully load your page in 30 seconds, real visitors on mobile connections are having a terrible experience.

Webflow — a website builder — having a homepage too heavy to analyze in 30 seconds is particularly notable.

What the Best Pages Do Differently

The pages that scored 78+ shared common patterns:

They front-load proof. Social proof appears above the fold, not buried at the bottom. Specific numbers and ratings, not just logos.

They pick one CTA and commit. One primary action, visually dominant, with benefit-driven copy. Secondary options exist but are clearly deprioritized.

They're specific about value. Instead of abstract taglines ("Where Finance Connects"), top scorers lead with concrete outcomes. What does your product do? Who is it for? What result will I get?

They offer low-friction entry points. Free tools, trials, or resources alongside the main conversion CTA. Multiple ways in, different commitment levels.

They load fast. The best-scoring pages consistently loaded under 2.5 seconds. Pages with 40+ images or heavy animations scored lower on performance, which drags down the entire experience.

What This Means for Your Landing Page

If you're running a SaaS product, the odds are your landing page has at least one of these three issues: a generic CTA that doesn't communicate value, insufficient social proof, or no low-friction capture path for visitors who aren't ready to buy yet.

The good news: these are all fixable without a redesign. Improving CTA copy, adding a testimonial with real numbers, and creating a simple lead magnet can move the needle significantly.

Want to see where your page stands?

Run a free landing page audit →

Leak Detector analyzes your page across all 8 categories in under 30 seconds and gives you a prioritized list of what to fix first.

---

Methodology: 48 SaaS landing pages were analyzed using Leak Detector's AI-powered audit system between January and February 2026. Each page was scored across 8 categories (Headline & Value Proposition, CTA, Social Proof, Forms, Visual Hierarchy, Trust Signals, Mobile, Performance) on a 0-100 scale. 40 pages returned full results; 8 timed out or blocked access. Scores reflect the state of each page at the time of analysis.