01
Decide what each event proves
A visitor opening a phone or WhatsApp link has shown an intention to contact you. They may close the app, abandon the message or call outside your operating hours. Recording that interaction is useful; labelling it a completed conversation or qualified lead overstates what the website can observe.
Write the event meaning before configuring Analytics. Google recommends generate_lead for submitting a form or a request for information, and separate lead-funnel events for later qualification. The relevant event should fire when that step actually occurs, rather than because a related button was clicked.
- Contact-link click: the link was activated.
- Successful form: the receiving service confirmed submission.
- Received enquiry: the team or connected system has the message.
- Qualified lead: someone checked the enquiry against agreed criteria.
- Customer: the actual commercial conversion is recorded.
02
Make an event map for your real links
For each contact action, record the page, link selector, destination type, event name, permitted parameters and trigger. A proposed WhatsApp event could be named click_to_whatsapp. This is a suggested naming choice, not an automatic GA4 event or a claim that this site already implements every WhatsApp variant.
This repository records click_to_call for tel: links. Its generic data-analytics-event hook supports explicitly named link interactions. If your site adds a WhatsApp link, wire the intended event to that actual element or an appropriate destination rule, and test it. Avoid attaching two handlers that count one action twice.
- Use page_path and a controlled link_location label to compare placements.
- Keep phone numbers, email addresses and message text out of Analytics parameters.
- Define how keyboard activation and dynamically inserted links are handled.
- Check every contact component, not only the homepage button.
03
Treat consent as part of the trigger
Project Monet's trackAnalyticsEvent helper checks whether pm_analytics_consent is granted before sending an event to gtag. Its form emits generate_lead only after both a successful HTTP response and success=true from the receiving service. This demonstrates the intended guard; it does not prove delivery to a specific business inbox or real-world lead quality.
Your implementation needs a clear consent state and a consistent way to apply it to every tracker. A click count gathered under one consent rule should not be compared casually with another period or tool that observes a different population.
04
Use a small acceptance worksheet
Test with identifiable test sessions and record the page, action, consent state, expected event, observed event and result. Use GA4 DebugView or the relevant development diagnostics to inspect the event that arrived. Do not leave test identifiers or personal information in public event parameters.
- Consent denied + contact click: no Analytics event under this site's policy.
- Consent granted + one call click: one call-interaction event.
- Consent granted + configured WhatsApp click: one event for that actual element.
- Form validation failure or provider error: no successful-submission event.
- Confirmed successful submission: one generate_lead event.
- One activation on a tagged contact link: no duplicate generic event.
Also check whether a rapid navigation loses an event and whether opening the destination still works. Tracking must support the contact journey. It should never delay or block a customer just to improve the dashboard.
05
Compare interactions with the team's records
Keep separate columns for contact clicks, confirmed submissions, received enquiries, qualified enquiries and customers. Where systems can be reconciled safely, use their actual records rather than assuming a fixed percentage of clicks becomes leads. If you cannot observe the later stage, leave it unknown.
Marking an event as a GA4 key event changes reporting treatment; it does not establish that a conversation happened. Explain what your chosen key event represents to anyone reading the report. Segment staff tests and spam where reliable evidence allows.
06
Choose measurement that the business can maintain
The calls and WhatsApp enquiry guide covers making the contact path clearer. This guide addresses the separate implementation question: what the resulting events can truthfully tell you.
Bring the event map and test worksheet to a website measurement review. Agree responsibility for the website, Analytics and the team's follow-up records before calling the system complete.
Sources
Primary and supporting sources
Use the linked documentation and labelled build examples to check the evidence. Business examples are planning tools; they do not establish customer outcomes or guarantee search visibility.