Conversion Rate Optimisation
Improves your website or app so that a higher percentage of visitors complete a desired action, such as a purchase or signup.
If you have ever run a test and thought it was conclusive after a few hundred visitors, you already know statistical significance is not a suggestion.
Key points
- Define your conversion goals and measure a baseline before making any changes.
- Focus CRO efforts on high-traffic or underperforming pages for the biggest upside.
- Use A/B testing to compare page elements, but ensure you collect enough data for statistical reliability.
- Combine behavioural analysis tools like heatmaps and session recordings with analytics for deeper insights.
- Avoid testing too many changes at once; isolate variables to know what caused the result.
- Regularly recalibrate benchmarks and targets as industry, device, and traffic conditions evolve.
A/B test on a product page lifts signups by 14%
An ecommerce site ran an A/B test on its product page CTA. The original button said 'Buy Now' and had a conversion rate of 2.1%. The variant changed the copy to 'Get Started' and moved the button above the fold. After 5,000 visitors per variant, the variant achieved a conversion rate of 2.4%, a 14% relative lift. The result was statistically significant at 95% confidence. The site implemented the change and saw consistent improvement over the next month.
Four common mistakes that waste your CRO efforts
- Skipping baseline measurement Optimising without a clear conversion baseline makes it impossible to measure improvement. Always define and record your current conversion rate first.
- Testing too many changes at once Changing multiple elements simultaneously confounds results. Isolate one variable per test to know what caused the change.
- Ending tests too early Judging a test before reaching statistical significance duplicates noise. Let tests run long enough for reliable data.
- Ignoring mobile users Mobile traffic often converts differently. Poor Core Web Vitals on mobile can hurt conversions, so test and optimise for mobile devices separately.
Four steps to set up a reliable CRO process
- Define your conversion goals Know which actions matter: purchases, signups, form fills, or other key events. Clear goals guide every test.
- Measure a baseline Calculate your current conversion rate with a reliable tracking setup. Without a baseline, you cannot validate improvements.
- Analyse user behaviour Use heatmaps, session recordings, and analytics to find friction points and drop-off areas on pages.
- Test and iterate Run A/B tests on high-traffic or underperforming pages. Use statistically valid samples and implement winning variations.
Four tools to include in your CRO toolkit
- Analytics platforms Google Analytics tracks conversions and user flows. Set up key events to measure specific actions.
- Behaviour analysis tools Heatmaps and session recordings (e.g., Hotjar) reveal where users click, scroll, and abandon journeys, helping you identify pain points that hurt conversion rates.
- A/B testing software Optimizely or VWO allow controlled experiments on page elements like CTAs, forms, and layouts, with built-in statistical analysis to ensure reliable results.
- Page speed monitoring Tools like PageSpeed Insights or Lighthouse measure Core Web Vitals, which directly affect conversion rates on mobile and desktop. Optimising these can reduce friction.
Four situations that change your conversion targets
- After a redesign A site redesign often changes user behaviour. Reassess conversion rates after launch to set new baselines.
- When privacy rules change Regulatory shifts such as GDPR or CCPA can affect cookie consent and tracking accuracy, impacting conversion measurement and requiring updates to your digital marketing analytics setup.
- When traffic sources shift Different traffic types (organic, paid, social) convert at different rates. Segment and recalibrate targets accordingly.
- After a major algorithm update Google algorithm updates can shift organic traffic quality, altering conversion patterns. Reassess targets after known updates.
Common questions
How long should you run an A/B test?
Run the test until you reach statistical significance, typically at least one full business cycle (e.g., one week) to account for daily variations.
What are conversion rate optimisation best practices?
Define goals, measure baselines, isolate variables in tests, use behavioural analytics, and let tests reach statistical significance before deciding.
How does conversion rate optimisation work?
CRO works by analysing user behaviour, forming hypotheses, running A/B tests, and iterating based on data-driven results.
What are the best CRO tools?
Top tools include Google Analytics for tracking, Hotjar for heatmaps, and Optimizely or VWO for A/B testing.
Sources
- Google Analytics documentation Official guidance on setting up conversion tracking and events
- Optimizely CRO glossary Definition and framework for CRO
- VWO CRO guide Step-by-step process and best practices