Conversion rate optimization (CRO) is the practice of improving a website or app to increase the proportion of users who complete a defined action, such as purchasing, signing up, or requesting a quote. Teams use research, analytics, and experiments to identify barriers and evaluate changes.
CRO helps teams get more value from existing traffic. Useful improvements make it easier for people to understand an offer, make a decision, and complete their task.
How do you calculate conversion rate?
For a user-based metric:
Conversion rate = users who complete the target action ÷ eligible users × 100.
Count each converting user once within the defined measurement window. Do not substitute total event count: one person can submit more than once. For a session-based metric, divide sessions with a conversion by eligible sessions instead. Keep the action, audience, denominator, and measurement window consistent when comparing results.
For a quote-request page, eligible users could be the users who reach that page. If 200 of 10,000 eligible users request a quote, the conversion rate is 2%. A rate of 2.5% represents a 0.5 percentage-point increase, or a 25% relative lift. These illustrative numbers show the size of a difference, not whether a change caused it or whether it is statistically reliable.
How does a CRO process work?
- Define success. Choose a primary metric and verify that events fire correctly. Add guardrails such as lead quality, refunds, or support requests.
- Find the barrier. Use funnel data to locate drop-off. Combine it with usability sessions, interviews, or support feedback to investigate why people struggle.
- Form a hypothesis. Connect an observed problem, a proposed change, and an expected outcome.
- Test and evaluate. Where traffic supports it, use a randomized A/B test. Set the evaluation method and stopping criteria in advance; interpret the effect alongside its uncertainty and guardrails.
- Apply what you learn. Release useful improvements, investigate inconclusive results, and monitor downstream outcomes.
A CRO example: a quote request form
Suppose usability sessions show that travelers abandon a quote form because they think submitting it commits them to a purchase.
The team tests a clear explanation beside the submit button: “Request a quote. No payment required.” This copy is appropriate only if it accurately describes the service.
| Decision | Example |
|---|---|
| Hypothesis | Explaining the commitment will help interested travelers submit the form. |
| Primary metric | Percentage of eligible users who submit a quote request. |
| Guardrails | Qualified leads per eligible user and completed bookings per eligible user. |
More submissions can increase workload without producing more qualified leads. Review those downstream results before calling the change a success.
What if traffic is too low for A/B testing?
Use usability testing and analytics to prioritize evident problems, such as broken validation or unclear requirements. Monitor changes after release, but treat before-and-after comparisons cautiously: traffic mix, campaigns, and seasonality can also affect conversion.