Sitesy offers lead generation and conversion rate optimisation services, giving website analytics a commercial brief: support decisions about qualified enquiries and customer acquisition. For developers, that means modelling real business states rather than treating every interface interaction as a lead.
Define events around confirmed outcomes
Separate form_start, enquiry_submitted and contact_click in the event plan. Fire enquiry_submitted only after the application confirms that it accepted the request, not when the visitor presses Submit. A telephone-link click should remain a contact intent event unless a call is independently confirmed.
Document each event's trigger, owner and permitted properties. An opaque enquiry identifier can help reconcile an accepted request with later lead records. Account for duplicate submissions and event retries so one enquiry does not inflate the apparent conversion total.
Connect web events to lead quality
Record qualification and sales outcomes in the lead-management system, then join them to enquiry records through a controlled reporting process. Keep rejected enquiries visible as a separate reporting state. Excluding them from the funnel would conceal the quality problems the business needs to diagnose.
Define each metric's denominator and date basis. Form completion rate could mean accepted submissions divided by form starts; qualification rate could mean qualified enquiries divided by all enquiries in that cohort. Consistent definitions matter more than a dashboard full of ambiguous percentages.
Test reliability before optimising the interface
Exercise validation failures, successful submissions, refreshes and duplicate event delivery in testing. Reconcile accepted form records against analytics counts, allowing for consent choices and blocking. A discrepancy is a diagnostic prompt, not automatic proof that the application lost leads.
Minimise collected data and apply UK consent and data-protection requirements to the tracking design. Keep names, email addresses and free-text enquiry content out of analytics payloads. An opaque identifier still needs appropriate safeguards if it can be linked back to a person.
Make the instrumentation answer a decision
Suppose mobile form starts rise while accepted submissions fall. Check browser errors and device-specific validation before redesigning the page. After a fix, compare similar periods and monitor qualified enquiries as well as submission counts; easier completion alone does not prove better lead quality.
Account for delayed qualification before declaring an experiment successful. A recent cohort may look weaker simply because sales staff have not reviewed it yet. Sitesy's commercial focus suggests the right acceptance question: does the evidence support a better acquisition decision, not merely a cleaner chart?
Explore this topic across our social channels
See the Sitesy discussion and related posts:
Top comments (0)