What counts as a conversion?
A conversion is an outcome you define in advance. A consultancy might choose an enquiry, a shop a completed purchase, and a newsletter a confirmed subscription. Give the event a name that describes exactly what happened.
Distinguish an intermediate step from the final outcome. Clicking through to a payment page does not prove payment. Clicking 'submit' may produce an error message. Record success after the action has been processed successfully.
| Goal | Measure when | What is not proof yet? |
|---|---|---|
| Enquiry | The form has been successfully processed. | Clicking the submit button. |
| Purchase | The order or payment is confirmed. | Opening the checkout. |
| Registration | Registration is complete under your process. | Visiting the registration page. |
Should you measure in the browser or on the server?
A browser event works when the website can confirm a successful action. Ensure that refreshing or reopening a thank-you page does not silently repeat the conversion. A consistent order or enquiry reference can help prevent duplicates.
A server event can suit outcomes confirmed later, such as a payment notification. You still need to establish the connection to a measured visit or campaign. Server measurement does not automatically recover missing source information.
How do you calculate a conversion rate?
Choose the denominator first. For a visit-based rate, divide visits with at least one selected conversion by total visits, then multiply by 100. Keep that definition consistent when comparing campaigns.
For example, if 20 of 1,000 visits produce at least one enquiry, the rate is 2%. If those 20 visits produce 24 enquiries, that is 24 events and 20 converting visits. This is a calculation example rather than a benchmark for your website.
How do conversions help you improve the website?
Compare outcomes by source, campaign and landing page. Then inspect the steps before conversion. A page attracting traffic without useful next actions presents a different question from a form that people open but seldom complete.
With little traffic, avoid drawing conclusions from one additional enquiry. Compare a suitable period and change one important element at a time. Keep form contents, names and email addresses out of analytics events: recording the outcome is usually enough.
Statora
How can Statora help?
Statora supports campaigns and conversion measurement from Light. You can record goals through browser events or a connection to your own server process. Outbound clicks and downloads provide additional signals, but are not automatically confirmed enquiries or purchases. Choose a plan and measurement point that match your actual goal.
Where to start
- Define one specific, confirmed outcome as your first goal.
- Test success, failure and repetition of the same action.
- Compare converting visits per campaign using consistent definitions and periods.