Tracking Leads in GA4: Forms, Calls and WhatsApp Clicks as Key Events

On a corporate or B2B website, nobody pays at checkout. The moments that matter are a quotation form sent, a tap on the phone number, or a tap on the WhatsApp button. If Google Analytics 4 is not told which of those actions count, every report treats a buyer who asked for a price list the same as a visitor who read one page and left, and any Google Ads campaign optimising on that data is guessing.

Below: what to mark, how to set it up, how to avoid double counting, how to feed Google Ads, and which source to trust when GA4 and the CRM disagree.

Key events in one paragraph

A key event, in Google’s words, is “an event that measures an action that’s particularly important to the success of your business”. Any event GA4 collects can be marked as one, and key events then show up as a column in standard reports such as Landing pages and User acquisition, and in the attribution reports (Google Analytics Help: About key events). Google Ads works with its own conversions, which you can create from GA4 key events. The two stay separate: Google states that Google Ads conversions do not appear in standard Google Analytics reports. In short: a key event becomes a conversion only when you import it into Google Ads.

What to mark as a key event, and what to leave as a plain event

The common mistake is marking too much. Scroll depth or downloads sitting next to quotation requests inflate the “lead” number until nobody can read it. A practical rule: mark something as a key event only if a salesperson would want to know it happened.

Keep forms, calls and WhatsApp as three separate key events instead of one combined “lead”. They have very different close rates, and if they are merged you can no longer see that, for example, one landing page produces WhatsApp taps that rarely reach a quotation.

Setting up forms, calls and WhatsApp clicks

Forms: count the success, not the attempt

GA4’s enhanced measurement can record form_start the first time a user interacts with a form in a session, and form_submit when the user submits it, with parameters such as form ID and destination (Google Analytics Help: Enhanced measurement events). Those parameters can only be used in reports after you create custom dimensions for them. Before you rely on form_submit as your lead count, test it on your real forms. Depending on how a form plugin or script sends its data, the automatic event may fire on attempts that fail validation, or not fire at all. The more dependable options are an event your developer fires only when the form plugin confirms success, or a dedicated thank-you page.

If you use a thank-you page, GA4 lets you create an event without code: in Admin, under Data display, open Events, choose Create event, base it on page_view and enter the thank-you page URL. You can switch on “Mark as key event” in the same step (Google Analytics Help: Create or modify key events). Make sure that page cannot be reached any other way, such as from the menu or search results, or it will count visits that are not leads.

Phone calls: measure the tap

Google Ads can track calls to the number on your website using Google forwarding numbers, which swap your number for a tracked one (Google Ads Help: Track calls to a phone number on your website). For most of this region, that option does not exist. At the time of writing, Google’s list of countries where forwarding numbers are available does not include Saudi Arabia, the UAE, Egypt, Qatar, Kuwait, Bahrain or Oman; the list changes from time to time, so check it before you plan around it (Google Ads Help: Google forwarding number). Where forwarding numbers are not supported, Google Ads reports only “call clicks”, the taps on the call button in the ad, and Google notes that calls from people who visit your website after seeing the ad are not included in that metric (Google Ads Help: Understanding call clicks).

In practice, a Gulf or Egyptian B2B site measures a website call as a tap on the tel: link, tracked as a custom event, for example through a Google Tag Manager click trigger on links that start with tel:, or an event your developer adds. Check in GA4’s real-time view that a test tap is recorded before you rely on it. A tap is an intention to call, not a connected call, so treat it as a strong signal rather than a confirmed lead.

WhatsApp: build on the outbound click

Enhanced measurement records an outbound click event each time a user clicks a link that leads away from your domain, with parameters including link_url and link_domain (Google Analytics Help). A standard WhatsApp link (such as wa.me or api.whatsapp.com) points to another domain, so in most setups it should be recorded this way, but Google’s documentation does not name WhatsApp, so confirm it with a test tap. Even when it is recorded, it is mixed in with every other outbound link: partner sites, social profiles and map links. To isolate it, create a custom event (for example whatsapp_click) based on the click event, with a condition that link_url contains the WhatsApp address your button uses, then mark only that new event as a key event. Check the exact URL your button produces first, because some buttons use wa.me and others api.whatsapp.com. For WhatsApp beyond the website, see click-to-WhatsApp ads for B2B leads. If your WhatsApp button is a widget that opens a chat without a normal link, enhanced measurement may not see it at all. In that case, a Tag Manager trigger or a developer-fired event is needed. Test the button yourself before trusting the report.

The same caution as for calls applies: a WhatsApp tap opens a chat window and does not guarantee a message was sent. Whether the chat becomes a lead depends on the first reply and the qualifying questions your team asks, which GA4 cannot see; see why WhatsApp inquiries go cold.

Stopping the same lead from being counted twice

Double counting has four usual causes.

1. Two events for one action. If the form fires a success event and the thank-you page is also a key event, every lead counts twice. Choose one and keep the other as a plain event or remove it.
2. The counting method. GA4 offers two settings per key event: “Once per event”, which counts every time the event fires, and “Once per session”, which counts it once in a session. Google’s example: if someone triggers a key event five times in a session, the first setting counts five and the second counts one. Once per event is the recommended default, but key events created from old Universal Analytics goals default to once per session. Changes apply to future data only (Google Analytics Help: About key event counting method). For WhatsApp and phone taps, where one visitor often taps two or three times, once per session usually gives the more realistic number. For forms, once per event is fine if each submission is a separate request.
3. Two tags in Google Ads. If Google Ads has its own conversion tag on the thank-you page and you import the same GA4 key event, the account has two conversion actions for one lead. Only one of them should be primary.
4. Thank-you page reloads. A refresh or a back-button visit fires another page view. A success event fired by the form avoids this.

Sending key events to Google Ads

To create Google Ads conversions from GA4, Google lists three prerequisites: link GA4 and Google Ads, turn on auto-tagging in Google Ads, and be an administrator or editor of the GA4 property. You then create the conversion from Conversion management in GA4’s Advertising section, choose the Google Ads account and select the key events to import (Google Ads Help: Create Google Analytics 4 conversions from Google Ads).

Three details matter for lead accounts:

• Imported conversions arrive as secondary. Google sets them to secondary “to prevent counting the same event twice for bidding”. They are reported but do not steer Smart Bidding until you make them primary in Google Ads. Decide on purpose which actions the campaigns should optimise for.
• Not every key event belongs in bidding. Making the quotation form primary and keeping phone and WhatsApp taps secondary is a sensible starting point when taps are numerous and loosely qualified. If your sales team confirms that WhatsApp chats close well, promote that action later.
• Attribution settings differ. For web conversions you choose whether Google paid channels only, or paid and organic channels, can receive credit. Google states that properties creating a link for the first time default to Google paid channels. Imported Analytics conversions also do not show view-through conversions. GA4 offers its own data-driven and last-click models (Google Analytics Help: Attribution), so the same lead can be credited differently.

When the website and ad account are run by different people, this is where setups break. Our website and performance marketing teams treat tracking as one job, and the ad accounts stay in the client’s name.

Why GA4 will never match the CRM, and which number to trust

Even a perfect setup will not make GA4, Google Ads and your CRM agree.

• Consent. With consent mode, when a visitor declines cookies, Google tags send cookieless pings instead of storing cookies. In the “advanced” implementation, tags load and send those pings, and GA4 uses behavioural and conversion modelling to fill gaps. In the “basic” implementation, tags are blocked until consent is given, and Google notes that only the advanced implementation gets behavioural modelling in Analytics (Google Analytics Help: About consent mode). Either way, some key events are modelled or missing, while your CRM holds every real inquiry. Our guide to cookie consent and consent mode on GCC websites covers the choice between the two.
• Different definitions. GA4 counts taps and submissions; the CRM counts deduplicated people and companies.
• Offline steps. Many B2B inquiries start with a call to a salesperson’s mobile, a referral or an exhibition meeting, and never touch the website.
• Different attribution. Google Ads, GA4 and the CRM each assign the lead to a source using their own rules and time windows.

So stop trying to reconcile the totals and give each source a job:

Watch the ratio, not the gap. To take a hypothetical example: if GA4 usually shows 120 form key events against 90 CRM inquiries, a sudden move to 120 against 40 means something broke: spam, tracking, or a form that stopped reaching the inbox. For how these numbers fit on one page for leadership, see the monthly marketing dashboard a CEO should actually read.

Closing the loop: tell Google which leads were real

Tap-level tracking tells Google Ads who showed interest, not who became a customer. Two tools help close that gap.

First, GA4’s recommended lead-generation events give you standard names for the stages after the inquiry: generate_lead when a form is submitted, qualify_lead when a lead meets your criteria, disqualify_lead, working_lead when a representative is in contact, and close_convert_lead or close_unconvert_lead for the outcome (Google Analytics Help: Recommended events). Using these names from the start makes CRM stages easier to add later; our CRM requirements checklist covers which stages to record.

Second, Google Ads offers enhanced conversions for leads, which Google describes as a way to measure offline leads using first-party data already stored in your CRM, hashed before it is sent to Google (Google Ads Help). Optimising for qualified leads instead of taps may lower lead volume while raising quality, the trade most B2B sales teams want.

Setup checklist

• Three separate key events: form success, phone tap, WhatsApp tap. Nothing else marked as key.
• One source per action (success event or thank-you page, not both).
• Counting method chosen on purpose for each key event.
• WhatsApp event filtered by link URL, tested on mobile and desktop.
• GA4 and Google Ads linked, auto-tagging on, primary conversions chosen deliberately.
• Consent mode implementation documented, so everyone knows which data is modelled.
• A monthly check of GA4 key events against CRM inquiries, with the ratio recorded.

Run the quick test from the double-counting section today, before changing anything else. If the totals come out right, record this month’s GA4-to-CRM ratio as your baseline; if they do not, fix the setup first, because every later report and bid depends on it.