How it works
Conversion rate relates a defined successful action to a defined population of visits or users. Both numerator and denominator need documentation. A contact request, purchase and newsletter sign-up are different outcomes; combining them can conceal whether the main task has improved.
A realistic example
If a service page receives 200 eligible visits and 10 valid enquiries, the visit-based enquiry rate is 5%. Duplicate events, spam or a different session definition would change that calculation. The rate says nothing by itself about the quality of the enquiry or eventual business outcome.
What to check
Verify that the success event fires once after actual completion. Compare device, source and page-purpose segments over a consistent period. Inspect form failures and abandoned steps alongside the headline rate, using only data collected under the applicable privacy choices.
Limits and next steps
A percentage based on few visits can fluctuate sharply. Do not infer that a new design caused every change. Keep definitions stable, report sample size and use task quality as well as volume when assessing an improvement.