The Conversion Myth
Most conversion rate advice on the internet is based on anecdote, not data. "Put the CTA above the fold." "Use orange buttons." "Remove the navigation." Some of this is directionally correct; most of it is dangerously context-dependent.
Here's what large-scale testing actually shows.
The Headline Is Everything
Unbounce research consistently shows the headline accounts for over 80% of the variance in conversion rate between similar page designs. Not the button color. Not the hero image. The headline.
A high-converting headline does one of three things: states a specific outcome ("Get 3× more qualified leads in 90 days"), names and solves a specific problem ("Stop losing clients to a website that doesn't convert"), or directly challenges a held belief ("SEO doesn't have to take 12 months").
Vague aspirational headlines ("Transform your business") perform poorly because they say everything and nothing simultaneously.
One Page, One Job
The highest-converting landing pages have a single call to action. Every link, every button, every navigation item that takes a user away from that primary CTA is a conversion leak.
Unbounce's analysis of 74 million landing page visits found that removing navigation from landing pages increased conversions by up to 100%. The navigation gives users an exit. Remove the exit.
Social Proof: Specificity Over Quantity
"Trusted by 10,000+ companies" is nearly worthless. "Helped a 3-person fintech startup generate £240K in revenue in six months" is gold.
The Baymard Institute's research on trust signals shows that specific, verifiable claims dramatically outperform generic ones. Named clients, measurable results, and verbatim testimonials (with full names and photos) are the most persuasive formats.
Form Length and Conversion Rate
Hubspot's research on 40,000 landing pages shows a direct inverse relationship between form field count and conversion rate - up to a point. Going from 11 fields to 4 fields increases conversions by 120%. But going from 4 to 1 doesn't double it again.
Ask only for what you need at this stage of the relationship. Email beats email + first name + last name + company + phone + budget.
The Hero Image Decision
Stock photos of smiling people in offices hurt conversions. This is consistent across tests. Real photos of your product, your team, or your actual customers consistently outperform generic imagery.
If you must use stock, choose images that are specific to a context, not aspirationally generic. A photo of a dashboard with real data is more persuasive than a photo of someone looking at a laptop.
Speed as a Conversion Variable
Google's research shows that a 1-second delay in mobile page load time reduces conversions by 20%. For every 100ms improvement, conversion rate improves by 1%. Landing pages that take more than 3 seconds to load lose more than half their visitors before the page is seen.
Optimise images (WebP, lazy loading), use a CDN, eliminate render-blocking scripts. Page speed isn't a dev problem - it's a revenue problem.
