BUILD SEARCH FOUNDATIONS
How to measure website enquiries without confusing clicks with leads
An enquiry measure should count the outcome you want, not just a button press. Define a suitable lead and a completed contact action before comparing a new website with the old one.
Write a counting rule
Decide what belongs in the count: a genuine project enquiry, a completed booking or a request for a quotation. Exclude authorised tests and obvious spam consistently. Keep support questions and duplicate messages separate so a higher inbox volume is not automatically reported as more potential customers.
Start with a simple record
Use date, enquiry type, fit, reported source and next step. Keep personal details in your existing authorised customer system rather than duplicating them in a general reporting sheet. If you ask how someone found you, allow “unknown” and do not present their recollection as precise tracking data.
Distinguish the stages
| Stage | What it establishes |
|---|---|
| Contact button clicked | Someone started an action |
| Email draft opened | An app may have prepared a message |
| Message received | The enquiry reached the business |
| Suitable project discussed | The enquiry fits the service |
Compare like periods
Choose comparable time windows and note promotions, seasonal changes and sales activity. Keep absolute numbers alongside percentages. A change from two to four suitable enquiries is useful to record but too small to establish a dependable conversion pattern.
Add tracking deliberately
If you later install analytics, define events around the actual workflow and verify they fire as intended. Review the relevant privacy and consent arrangements first. Do not call an email-link click a submitted form. Record what is and is not measured.
Review whether the site helps suitable customers understand the offer and send enough information for the next conversation. Use recurring questions to improve service content. The objective is a clearer business process and useful evidence, not a dashboard that assigns precise attribution where none was collected.
Practical guidance with illustrative examples. Read our editorial standards or send a correction.