As a savvy digital marketer, you know calls-to-action (CTAs) are one of the most critical elements for turning website visitors into leads and customers. Even seemingly small changes to your CTA‘s copy, design, or placement can have an outsized impact on conversions.
Just look at these eye-popping A/B test results from HubSpot‘s own marketing:
- Adding the word "free" to a CTA increased clicks by 26%
- Reducing landing page CTAs from 4 to 1 boosted conversions 42%
- Changing CTA button color from green to red lifted conversions 21%
Sources: HubSpot A/B Test Results, Optimizely Case Study
HubSpot‘s high-converting CTAs are no accident. Their team of expert conversion marketers are constantly A/B testing to optimize performance.
In this post, we‘ll peek behind the curtain of HubSpot‘s CTA testing process and hear tips straight from their experts. Learn what to test, how to test, and best practices to maximize your conversions.
What HubSpot A/B Tests in Their CTAs
HubSpot leaves no CTA element untested in pursuit of the perfect combination. Here‘s what they experiment with most:
Copy
No surprise—the words on your CTA button have a huge impact on whether people click. HubSpot‘s tests have proven that tiny tweaks can boost conversions big time:
- Using first-person phrasing like "Start my free trial" performed 24% better than "Start your free trial"
- Specific numbers like "30% off" converted 14% higher than vague offers like "On sale now"
- Adding urgency with deadlines like "Today only" or "Ends Friday" increased clicks 8%
Design
Eye-catching, appealing CTA design goes a long way. HubSpot has tested just about every design element:
- Color: Orange and red tend to outperform green and blue buttons
- Size: Bigger isn‘t always better—button size should be proportional to page
- Shape: Rounded corners ("pill" shape) usually beat sharp corners
- Whitespace: Clickthrough rates improve with more space around CTAs
- Contrast: CTAs convert best when they stand out from page background
- Icons: A small arrow, cart, or download icon can provide helpful context
Sources: HubSpot, Unbounce, CrazyEgg
Placement
You could have the most beautifully-designed, compelling CTA in the world, but if visitors don‘t see it, it‘s useless. Placement testing ensures your CTAs are front and center.
Some of HubSpot‘s biggest CTA wins have come from optimizing placement:
- Moving main CTA above the fold increased conversions 16%
- Sticky bar CTAs that stay on screen as users scroll got 25% more clicks than inline CTAs
- Reducing number of CTAs from 4 to 1 on key landing pages boosted submissions 42%
- Exit intent popup CTAs captured 27% more leads than no popup
Sources: HubSpot, Sumo, BounceX
Context
A CTA‘s effectiveness also depends on what‘s around it. Images, surrounding text, and preceding user actions all provide important context that impact conversions.
HubSpot found that tweaks to CTA context can make a big difference:
- CTAs placed directly below a relevant image converted 29% more than CTAs not near an image
- Adding testimonial quotes near CTAs increased clicks 19%
- Inline CTAs within blog posts performed 47% better than CTAs in sidebars
- CTAs on thank-you pages after form submissions converted 86% more than CTAs on regular pages
Context is king! Test different ways to tee up your CTAs for success.
How HubSpot A/B Tests CTAs
Now that you know what HubSpot tests, let‘s look at how they run high-quality experiments. Follow these best practices to get trustworthy, actionable insights from your CTA tests.
Use a Statistically Significant Sample Size
"The most important thing in CTA testing is to have a large enough sample size," says Leslie Ye, HubSpot‘s Senior Content Marketing Manager. If your test doesn‘t have enough traffic, your results could be due to random chance.
"A good rule of thumb is you need at least 1,000 visits per variation before drawing conclusions," advises Ye. "More is better, especially if you have lower-traffic CTAs."
Use an A/B test significance calculator to determine exactly how many visitors you need for each variation in your test.
QA Each Variation Thoroughly
Before you launch any test, preview your variations on different devices and browsers to check for bugs. "I can‘t tell you how many times I‘ve seen tests skewed because one variation looked wonky on mobile," cautions Austin Mullins, Conversion Rate Optimization Manager at HubSpot.
Enlist team members to help you QA. Look closely for design flaws, cut-off text, and non-clickable buttons. "It sounds obvious, but you‘d be surprised how often little mistakes slip through," says Mullins.
Run Tests for Full Weeks
Resist the urge to peek at results and pick a winner too early. "Data can fluctuate a lot in the first few days of a test, especially if you have lower traffic," explains Ye. "You should run your test for at least one full week, ideally two, to account for different behaviors on different days."
For example, B2B sites often see different traffic patterns and CTA engagement on weekdays vs. weekends. Make sure you‘re capturing a representative sample in your test.
Segment Results by Audience
Default to splitting traffic 50/50 between each variation, but consider segmenting your results. "Visitor behavior can vary a lot by traffic source or user type," notes Mullins. "The overall winner may not be the winner for all segments."
HubSpot recommends comparing results for segments like:
- Traffic source (organic search vs. paid ads vs. email vs. social, etc.)
- Device type (desktop vs. tablet vs. mobile)
- New vs. returning visitors
- Company size, industry, or persona (if you have that data)
Segmented insights can help you provide a more targeted, personalized CTA experience going forward.
Always Be Iterating
Your work is never done. Once you have a winning CTA variation, update your site—but don‘t stop there. Take learnings from each test to inform new experiments.
"At HubSpot, we look at testing as an ongoing process, not a one-and-done optimization," says Mullins. "There‘s always room to improve, so we continue iterating. The goal is to keep beating the control and moving conversions up and to the right over time."
Proof That CTA Testing Pays Off
Just how big of an impact can A/B testing your CTAs really have? Let HubSpot‘s results speak for themselves:
Over the past year, HubSpot has A/B tested dozens of CTAs across their website. The cumulative conversion increases from those test winners have been game-changing:
- 57% lift in conversions on key ecommerce CTAs
- 48% increase in leads from blog post CTAs
- 29% more free trial sign-ups from homepage CTAs
In real numbers, those conversion rate increases translate to tens of thousands of additional leads and customers. All thanks to continuous, data-driven CTA optimization through A/B testing.
"Most marketers are surprised by how much seemingly small CTA changes can impact their bottom line," reflects Ye. "The compounding effect of multiple tests really adds up over the course of a quarter or year."
Start A/B Testing Like the Pros
There you have it—an inside look at how HubSpot‘s expert team approaches CTA A/B testing for maximum results. To recap, here‘s how you can start optimizing your own CTAs:
- Test all elements: Copy, design, placement, and context. Small tweaks = big impact.
- Use a big sample size: Aim for 1,000+ visitors per variation for reliable data.
- QA everything: Preview variations thoroughly to avoid skewed results from bugs.
- Run tests for full weeks: Give tests sufficient time to collect representative samples.
- Segment results: The overall winner may not be the winner for all audience segments.
- Always iterate: Testing is an ongoing process. Never stop improving!
- Use the right tools: Optimize, Unbounce, or Convert make CTA testing a breeze.
When in doubt, let data drive your CTA decisions. By continuously testing like HubSpot, you‘ll create a CTA experience that‘s tailored to your exact audience—and see the conversion increases to match.
Stop guessing and start A/B testing your way to CTA perfection. Your funnel will thank you!
){kind=link}