Measurement · Insight

GA4 vs Google Ads Conversions: Which Should You Optimise Toward?

By Iyke Abel · Published 15 September 2026 · Updated 21 September 2026

When the same enquiry appears in GA4 and Google Ads, it is tempting to treat the two records as interchangeable. The useful question is whether each action represents the outcome you want the campaign to find.

Start with the business action

Write down what counts as success: a successful enquiry, a qualified lead, or a completed sale. A button click and a submitted form are different behaviours. Renaming both “lead” does not make them equivalent.

Then inspect how each version is collected. Check its trigger, counting settings, value, attribution settings, and the point in the journey at which it fires. Investigate differences before choosing a bidding signal.

Separate observation from optimisation

Google Ads can use conversions created from Analytics events as well as Google Ads conversion measurement. Primary actions are used for bidding when their standard goal is selected for the campaign. Secondary actions are generally for observation; custom goals are an important exception because their included actions can be used for bidding regardless of that setting.

Review the campaign’s actual goals, rather than assuming an action’s label tells the whole story. If two actions measure the same enquiry, avoid unintentionally asking the campaign to optimise toward both copies.

Choose the signal you can validate

There is no useful universal rule that one collection method is always right. Choose a validated action with the coverage and business meaning your campaign requires. Keep another measurement view where it helps diagnose gaps, and document why totals can differ.

For lead generation, the next improvement may be qualification feedback. A reliably measured enquiry still does not tell you whether the person became a customer.

Compare equivalent actions first

Write down the collection method for each action: a Google Ads conversion tag, an imported GA4 event, or feedback from another business system. Then check the actual business definition. Two records labelled “lead” might describe different points in the journey.

  • Collection: what event or source creates the record?
  • Counting: what happens when the same person repeats the action?
  • Timing: which reporting date, time zone and available processing period are being compared?
  • Attribution: which model, window and eligible interactions influence credit?
  • Campaign use: which goal is selected, and which actions actually contribute to bidding?

Align the comparison before calling a difference a tracking error. Agreement is not proof of accuracy, and disagreement is not proof that either platform is broken.

Example: one form, two conversion actions

Suppose a Google Ads tag and a GA4 import both record the same accepted enquiry. This is an illustrative setup, not a client result. If both copies are included in campaign optimisation, the campaign may receive two signals for one business action.

Validate each action, review the campaign goal and choose the intended optimisation signal. Keep a diagnostic view where useful, but check the actual campaign configuration rather than relying on the label “secondary.” Google’s documented custom-goal exception matters here.

Next, consider whether the enquiry is the best available business signal. Qualification feedback may be more useful once it is reliable and supported by the business process. That is a separate implementation decision, not a reason to switch goals without validation.

For a practical starting point, request a GA4/GTM audit. For the later stages, read how leads connect to revenue.