Consent Mode v2 introduces a revised framework for controlling Google Ads measurement in scenarios where users withhold consent for advertising or analytics storage. The update extends beyond the addition of two parameters, establishing a clearer distinction between the permissions to store advertising identifiers, transmit advertising-related user data, and utilise data for personalisation purposes.
For organisations within the UK or European Economic Area (EEA), Consent Mode v2 establishes a direct integration between the consent management platform (CMP), Google Tag Manager or the Google tag, Google Analytics 4, and Google Ads. As a result, a technically robust implementation is required to ensure both compliance with privacy governance requirements and the integrity of conversion measurement.
Consent Mode v2 does not guarantee comprehensive advertising measurement in all scenarios. It allows Google tags to modify their behaviour based on consent status and facilitates modelling where eligibility criteria are met. However, it does not reinstate user-level attribution in the absence of consent, substitute for a compliant CMP, or establish a legal basis for personal data processing.
1. The four consent signals that affect Google Ads measurement
Consent Mode v2 uses four principal signals for web measurement:
The first two signals were present in the original Consent Mode implementation. Consent Mode v2 adds ad_user_data and ad_personalization.
This distinction is important. A user may consent to advertising storage while not consenting to the use of user data for advertising purposes. Conversely, a user may allow certain measurement activity while declining personalised advertising.
Google states that ad_user_data is required for measurement use cases such as enhanced conversions and tag-based conversion tracking. The Google Ads consent mode reference also confirms that both ad_user_data and ad_personalization influence the availability of advertising features.
It is recommended to explicitly map each CMP category to the four Consent Mode v2 signals. Reliance on a single, general-purpose advertising consent value should be avoided unless the CMP documentation provides confirmation that the mapping is both accurate and auditable.
2. What happens when consent is granted or denied
When consent is granted
Where the relevant consent signals are set to granted, Google tags can operate with greater measurement capability.
Depending on the specific combination of signals, this may allow:
- Advertising cookies and identifiers to be read or written.
- Analytics cookies to support session and attribution continuity.
- Enhanced conversions to use consented user-provided data.
- Click identifiers such as GCLID or DCLID to support conversion attribution.
- Google Ads audiences and remarketing features to receive eligible data.
- Conversion data to be used by Google Ads bidding systems.
Feature availability depends on correct tag configuration, appropriate account linking, accurate conversion action settings, data quality, and the current consent state. Features are not enabled automatically solely by the presence of granted consent signals.
For instance, a status of ad_storage = ‘granted’ alone does not authorise the use of user data for advertising purposes; ad_user_data must also indicate the appropriate consent. Likewise, if ad_personalization = ‘denied’, personalised advertising is restricted even if other advertising-related consent signals are granted.
When consent is denied
When advertising or analytics storage is denied, Google tags must not read or write the relevant cookies. Depending on the implementation, tags may still send cookieless pings containing limited information for modelling and aggregated measurement.
Google describes these pings as a way to help its systems estimate conversion activity without using cookies to identify or track an individual across websites or applications. Google Ads consent mode modelling documentation explains that these signals can support conversion modelling where eligibility requirements are met.
The measurement consequences remain material:
- User journeys cannot be observed in the same way as consented journeys.
- Conversion paths may not be directly linked to ad interactions.
- Enhanced conversions are restricted when ad_user_data is denied.
- Personalised advertising and remarketing are restricted when ad_personalization is denied.
- Google Ads reporting may contain a mixture of observed and modelled conversions.
- Low-volume accounts may not qualify for conversion modelling.
Google states that modelling eligibility includes correctly implemented Consent Mode and a threshold of approximately 700 ad clicks over seven days for each country and domain grouping. This should be treated as an eligibility condition, not as a guarantee of complete recovery.
Distinguish between observed conversions, modelled conversions, and consent-related losses when evaluating performance. A decrease in directly attributed conversions does not inherently indicate a tracking defect; however, any unexplained reduction should prompt a review of the implementation and quality assurance processes.
3. Basic and advanced Consent Mode
Consent Mode can be implemented using a basic or advanced approach.
Basic Consent Mode
With the basic approach, Google tags are prevented from loading until the user interacts with the consent banner. If consent is denied, Google tags do not send data.
This approach constitutes a conservative data-collection model and may reduce the volume of measurement data available for modelling, as cookieless pings are not transmitted prior to consent being granted.
Advanced Consent Mode
With the advanced approach, Google tags load with a default denied state. Where consent remains denied, tags adjust their behaviour and may send cookieless pings. If consent is granted, the consent state is updated and the tags can oAdvanced Consent Mode can supply Google with additional information for modelling purposes, while continuing to restrict the use of advertising and analytics cookies in the absence of consent. Implementation requires a thorough assessment of the data transmitted, the CMP configuration, and the organisation’s privacy documentation.
The implementation order is critical. Google’s developer documentation for Consent Mode specifies that the default consent state should be set before commands that send measurement data. Consent updates should then be sent when the user makes or changes a choice.
The consent state should not be assumed to persist because of Consent Mode itself. Persistence is normally managed by the CMP or another approved consent storage mechanism.
4. What Consent Mode v2 does not solve
Consent Mode v2 is a technical control layer. It is not a complete privacy or measurement solution.
It does not:
- Replace a CMP or consent banner.
- Establish a lawful basis for storing cookies or processing personal data.
- Make non-essential cookies permissible before consent.
- Guarantee deterministic attribution for users who deny consent.
- Resolve discrepancies caused by incorrect conversion definitions.
- Correct broken redirects, lost click identifiers, cross-domain configuration errors, or duplicate tags.
- Guarantee eligibility for Google’s modelling systems.
- Ensure that all vendor tags respect the same consent state.
- Provide a complete audit trail of consent without appropriate CMP configuration and governance.
The ICO guidance on cookies and similar technologies states that organisations must explain what cookies do, why they are used, and obtain consent unless an exemption applies. Consent must be actively and clearly given. The guidance also makes clear that the rules apply to similar technologies used to store or access information on a user’s device.
Consent Mode serves as a technical mechanism for enforcing consent choices. It does not supersede the legal and governance obligations established under PECR or the UK GDPR.
5. Additional measurement controls
Consent Mode v2 can be integrated with supplementary Google measurement controls; however, each control requires independent validation to ensure it operates correctly.
URL passthrough
Where advertising storage is denied, URL passthrough can help retain certain ad-click information as users move between pages on the same domain. Google identifies parameters such as GCLID, DCLID, GCLSRC, _gl and WBRAID as relevant to this functionality.
URL passthrough requires:
- Correct Consent Mode implementation.
- A Google tag on the relevant pages.
- Redirects that preserve the required parameters.
- Site functionality that is not affected by query-string values.
- Analytics configuration that prevents these parameters from polluting page reporting.
Ads data redaction
ads_data_redaction can further limit the use of ad-click identifiers when ad_storage is denied. This can support a more restrictive privacy configuration, but may also reduce the information available for measurement.
The appropriate configuration is determined by the organisation’s privacy requirements and measurement framework. This setting should not be enabled or disabled without a comprehensive understanding of its implications for Google Ads reporting.
6. A practical implementation and QA approach
A controlled implementation process should encompass the following stages.
6.1 Document the consent model
Record the CMP categories, purposes, geographic rules, default states, retention periods, and vendor relationships. The documentation should identify which CMP choices map to each Consent Mode signal.
6.2 Set the default state before measurement
Apply the default values before Google Analytics, Google Ads, or other measurement commands send data. Region-specific defaults should be used only where they reflect an approved privacy and compliance position.
6.3 Implement consent updates
Each relevant user choice should generate an update to the four signals. The same process should support later changes, including withdrawal of consent.
6.4 Validate tag behaviour
Testing should cover:
- First page load before banner interaction.
- Acceptance of all categories.
- Rejection of advertising and analytics categories.
- Partial consent choices.
- Withdrawal of previously granted consent.
- Multiple domains and subdomains.
- Landing pages containing GCLID or other click identifiers.
- Single-page applications and route changes.
- Mobile and desktop implementations.
- Google Tag Manager preview and browser network requests.
6.5 Reconcile the reporting impact
Reconcile Google Ads conversions with source systems, GA4 key events, transaction databases, and consent rates. This reconciliation must consider attribution scope, conversion windows, time zones, deduplication, modelling, and reporting latency.
TagDataTrust’s analytics data quality service and Google Tag Manager implementation service address the wider implementation and QA issues that Consent Mode alone cannot resolve.
7. Conclusion
Consent Mode v2 should be regarded as a single control within a broader measurement and privacy framework. While it can enhance the management of consent-dependent measurement in Google Ads, it does not eliminate the requirement for a compliant CMP, rigorous implementation, or ongoing quality assurance. Organisations are advised to evaluate both the measurement benefits and reporting limitations prior to interpreting post-implementation outcomes.
If you’d like to hear more on this topic and set up a 1:1 call with us, please use the contact form. Contact a Consultant
