Design thinking resource

A/B Testing

A method used to compare two variations of a product, service, message or experience in order to determine which option performs more effectively.

Replace assumptions with evidence

Teams often have different opinions about which solution, feature or design is most effective. A/B testing helps reduce bias by generating evidence based on real user behaviour rather than personal preferences.

Instead of debating alternatives, two versions are presented to separate groups of users under similar conditions. By comparing outcomes, teams can determine which option better supports the desired objective.

This approach encourages evidence-based decision-making and helps organisations focus on measurable outcomes rather than assumptions.

Measure what matters

Effective A/B testing begins with a clear understanding of success. Whether the goal is improving engagement, increasing adoption, reducing errors or enhancing customer satisfaction, teams must identify meaningful metrics before testing begins.

The value of A/B testing lies in its ability to isolate specific changes and evaluate their impact. By focusing on measurable outcomes, organisations gain clearer insights into what influences user behaviour.

Well-designed experiments help reveal whether changes genuinely improve performance or simply appear promising without delivering meaningful results.

Support continuous optimisation

A/B testing is most effective when viewed as part of an ongoing learning process. Individual tests may produce small insights, but these insights accumulate over time and contribute to significant improvements.

Organisations that embrace experimentation often develop a deeper understanding of user behaviour and become more confident in their decision-making. Rather than seeking perfect solutions, they continually optimise and refine experiences based on evidence.

This mindset supports innovation, learning and long-term improvement.

Put the method to work

Put A/B Testing into practice

Define a clear objective

Identify what you are trying to improve and determine how success will be measured.

Create two alternative versions

Develop Version A and Version B, ensuring that only the elements being tested differ.

Collect and compare data

Expose representative users to each version and measure performance against the chosen metrics.

Apply the findings

Select the stronger-performing alternative and use the results to inform future improvements.

Why it matters

What A/B Testing makes possible

Improves decision-making

Testing generates evidence that helps teams choose between competing alternatives with greater confidence.

Reduces uncertainty

Rather than relying on assumptions, organisations can evaluate options using observable results.

Supports optimisation

Continuous experimentation helps improve products, services and customer experiences over time.

Encourages a culture of learning

Teams become more comfortable testing ideas, gathering evidence and refining their thinking.

Related concepts

Continue your learning

Continue exploring

Return to Execute and Test

Place this method back within the broader 5E Design Thinking Framework and continue developing your understanding.

View the stage →