Introduction

Marketing measurement used to be built around a relatively simple assumption: a visitor arrives, a tag fires, a cookie or identifier connects the journey, and the resulting conversion appears in the reporting stack. That assumption is no longer reliable.

Consent requirements, browser restrictions, changing Google controls, consent management platforms, GTM implementations and server-side architectures can all affect what is observed, what is modelled and what eventually appears in Google Ads, GA4 and CRM reporting.

The important question for B2B marketers is therefore not simply, “How many conversions did we record?” It is: “How much of the customer journey can we actually observe, how much is modelled, and can we trust the numbers well enough to make budget decisions?”

This guide explains the measurement changes marketers need to understand in 2026, with a practical focus on Consent Mode, Google Signals, GA4, Google Ads, GTM and regional consent.

The Big Shift: Measurement Is Becoming Consent-Aware

Consent is no longer something that sits beside measurement. It can directly influence how tags behave, what information can be collected, which identifiers can be used and how conversion gaps are handled.

Google's current Consent Mode documentation explains that consent signals are attached to data sent to Google and that consent mode can support modelling where direct measurement is incomplete. Advanced Consent Mode can send cookieless pings when consent is denied, while full measurement data is sent when the user grants the relevant consent. citeturn0search5turn0search6

This creates an important distinction: a lower reported conversion count does not automatically mean fewer real conversions. Some of the difference can come from reduced observability. At the same time, modelling is not a licence to assume that every missing conversion actually happened. The measurement layer needs to be understood before performance conclusions are drawn.

What Changed in Google Signals in 2026?

Starting June 15, 2026, Consent Mode within Google Ads became the single control for Google Ads cookie and identifier collection, while Google Signals continues to control the association of Google Analytics data with signed-in user information for behavioral reporting.

This distinction matters because Google Signals and Consent Mode are not interchangeable. They serve different purposes and should be treated separately when reviewing your measurement setup.

• Consent Mode is increasingly central to how Google advertising-related data collection responds to user consent.

• Google Signals remains relevant to Analytics behavioral reporting and association with signed-in Google information.

• Google Ads measurement should therefore be reviewed through the Google Ads consent framework rather than assuming the old Google Signals configuration controls everything.

For teams that built their measurement architecture years ago, this is a good reason to document exactly which setting controls which data flow rather than relying on historical assumptions.

Consent Mode and Google Ads Measurement

Consent Mode allows Google tags to adjust their behavior based on the consent choices made by website visitors.

For current implementations, Google uses four key consent signals:

• ad_storage — controls advertising-related cookies and storage

• analytics_storage — controls analytics-related cookies and storage

• ad_user_data — controls whether user data can be used for advertising purposes

• ad_personalization — controls whether data can be used for personalized advertising

For measurement, ad_user_data is particularly important because it affects advertising measurement use cases such as Enhanced Conversions and Google Ads conversion tracking.

A practical implementation should answer four key questions:

• What is the default consent state before the visitor makes a choice?

• When does the consent state update?

• Which Google tags are allowed to operate under each consent state?

• Can the implementation be consistently verified across countries, devices and page templates?

Google recommends setting the default consent state before tags or other code attempt to use or update consent. Your Consent Management Platform (CMP) should then communicate the visitor's consent choice to Google. Regional consent settings can also be configured where different requirements apply.

The key takeaway: Consent Mode is not simply a banner configuration. It directly affects how measurement tags behave and, ultimately, how much marketing data your platforms can observe.

Why Consent Can Change Reported ROAS

ROAS is a ratio. If the reported conversion value changes while spend remains constant, ROAS changes. That sounds obvious, but the reason behind the conversion change is often overlooked.

Imagine a campaign spends £10,000 and reports £40,000 in attributed revenue. Reported ROAS is 4.0x. If a measurement change causes only £30,000 to be directly observed, the reported ROAS becomes 3.0x even if the underlying commercial activity did not fall by 25%.

The opposite can also happen. A tagging change, duplicated event, changed attribution configuration or CRM import can inflate recorded conversions and make ROAS look stronger than the underlying business reality.

This is why performance teams should separate three concepts:

• Observed conversions: conversions directly recorded by the measurement system.

• Modelled conversions: conversions estimated to fill measurement gaps.

• Business conversions: outcomes confirmed in the CRM or other source of truth.

The three numbers can legitimately differ. The problem begins when they are treated as identical.

Consent-Aware GA4 Implementation

A reliable GA4 implementation starts with architecture, not reporting.

Before looking at dashboards, document the relationship between the CMP, consent signals, GTM, Google tags, GA4, Google Ads and the CRM. Every conversion should have a clear path from user action to recorded event to business outcome.

• Define the consent categories and their purpose.

• Set the correct default consent state before measurement tags execute.

• Send consent updates immediately after the user makes a choice.

• Use Google tags with built-in consent checks where applicable.

• Review custom GTM tags and add appropriate consent settings where required.

• Keep conversion event names and parameters consistent.

• Test the implementation before and after accepting or rejecting consent.

• Validate that regional rules produce the intended behavior.

Google's Google tag, Google Analytics, Google Ads, Floodlight and Conversion Linker tags include built-in consent checks. Custom tags that do not support these checks may require additional consent settings in Google Tag Manager.

The Hidden Problem: Double-Counting Conversions

One of the most dangerous measurement problems is not missing data. It is duplicated data.

Double-counting can appear after a CMP or GTM migration when the same conversion is sent through more than one route. For example, a form submission might trigger a Google Ads conversion tag directly, send a GA4 event that is imported into Google Ads, and also be sent through a server-side implementation.

The result can be a dashboard showing more conversions without anyone deliberately trying to inflate the numbers.

Before and after any CMP, GTM or server-side change, audit:

• Google Ads conversion actions

• GA4 key events

• Imported GA4 conversions in Google Ads

• GTM tags and triggers

• Server-side event forwarding

• CRM-to-ad-platform offline conversion imports

• Enhanced conversion implementations

• Form thank-you page and event-based triggers

The goal is simple: one business action should have a clearly defined measurement path, with documented rules for which system counts it.

Testing Consent by Region

A global website should not be tested only from one laptop in one country.

Google provides regional consent configuration capabilities, and its current guidance specifically discusses adapting consent defaults by visitor region. citeturn0search3

For a B2B business targeting the UK, Germany and the US, create a repeatable regional test plan.

• Open the site from a UK test environment.

• Open the site from a Germany/EEA test environment.

• Open the site from a US test environment.

• Check the initial consent state before interaction.

• Accept relevant consent categories and confirm the update.

• Reject relevant categories and confirm the expected restrictions.

• Inspect network requests and tag firing.

• Verify GA4 events and Google Ads conversion behavior.

• Repeat the test after CMP or GTM releases.

Do not assume that a banner displaying correctly means the measurement implementation is correct. The banner, consent signal, tag behavior and downstream reporting all need to agree.

UK vs EU/Germany: What Marketers Need to Consider

This article is not legal advice, and privacy compliance should be reviewed with appropriate legal or privacy specialists. From a measurement perspective, however, regional differences matter because consent requirements can affect how tracking technologies operate.

In the UK, the ICO's April 2026 guidance on Storage and Access Technologies covers cookies, tracking pixels, link decoration, local storage, fingerprinting, scripts and tags under PECR. It also explains where UK GDPR may apply when personal data is involved.

For EEA traffic, Google states that advertisers using relevant Google tags or SDKs for measurement and advertising features need to collect applicable consent and share consent signals with Google.

For marketers, the practical takeaway is simple: don't assume that the same consent setup will work identically across every market. Build your measurement architecture so regional consent states can be tested, documented and validated.

Server-Side Tracking vs Consent-Aware Client-Side Tracking

Server-side tracking is often presented as the answer to browser and measurement limitations. It can improve control over data flows, reduce some client-side complexity and provide a more structured measurement architecture.

But server-side tracking does not remove the need to respect consent.

If the underlying collection or processing requires consent, moving the request from the browser to a server does not automatically make the consent requirement disappear. The server-side architecture needs to receive and enforce the appropriate consent state.

The better question is therefore not “client-side or server-side?” It is:

“Which architecture gives us reliable measurement while preserving the user's consent choices and maintaining a documented, auditable data flow?”

• Client-side tracking can remain effective when implemented with correct consent controls.

• Server-side tracking can improve control and architecture, but consent signals still need to be propagated and respected.

• Hybrid architectures can be useful when different data sources have different measurement requirements.

• CRM and offline conversion data can provide an important business-level validation layer.

How to Measure the Cost of Lost Consent

“Lost consent” should not be treated as an unexplained percentage in a dashboard. It should be treated as a measurable impact on your marketing data and business reporting.

Start by measuring the percentage of visitors who provide the relevant consent. Then compare observed conversions from consented traffic with overall traffic, while keeping regional, campaign and channel differences visible.

A practical framework is:

• Measure the percentage of users who provide the relevant consent.

• Measure the conversion rate among users who have provided consent.

• Compare observed conversions with eligible modelled conversion estimates where available.

• Compare platform-reported conversions with CRM outcomes.

• Estimate the potential revenue or pipeline represented by unobserved conversions, clearly identifying it as an estimate rather than confirmed revenue.

Google provides Consent Mode impact reporting when an implementation has sufficient data and meets the relevant requirements. Modelling can help recover some measurement gaps, but the extent of that recovery depends on factors such as consent rates and implementation quality.

For B2B companies, the most useful measure may not be ecommerce revenue. It could be MQLs, SQLs, opportunities, customer acquisition cost or pipeline value.

The goal is not to create a perfect estimate. It is to understand how much of your marketing performance you can actually observe — and how much may remain invisible because of consent.

How to Audit Whether Your Marketing Data Can Still Be Trusted

A measurement audit should answer five questions.

1. Is the data collection technically correct?

• Are consent defaults set before tags execute?

• Do consent updates fire correctly?

• Are tags firing only when permitted?

• Are conversion events firing once?

2. Is the platform configuration consistent?

• Are Google Ads conversion actions documented?

• Are GA4 key events clearly defined?

• Are imported conversions duplicated?

• Are attribution settings understood?

3. Is regional behavior understood?

• Does the UK implementation behave as intended?

• Does Germany/EEA traffic receive the intended consent treatment?

• Does US traffic follow the intended configuration?

• Have real-world tests been documented?

4. Can platform data be reconciled with the business?

• Do Google Ads conversions reconcile with GA4?

• Do GA4 conversions reconcile with CRM submissions?

• Can MQLs and SQLs be traced back to campaigns?

• Are offline conversion imports documented?

5. Can management understand the limitations?

• Which numbers are observed?

• Which numbers are modelled?

• Which numbers come from CRM?

• Where are the known measurement gaps?

• What changed compared with the previous period?

What B2B Marketing Teams Should Change

The biggest change is cultural: measurement should no longer be treated as a dashboard problem owned only by the analytics or performance marketing team.

Marketing, analytics, web, privacy and sales operations should agree on what constitutes a conversion and which system is the source of truth.

• Document the measurement architecture.

• Create a consent and tagging change log.

• Run regional measurement tests after major releases.

• Keep a single conversion dictionary.

• Reconcile platform conversions with CRM outcomes.

• Separate observed, modelled and business-confirmed results.

• Explain measurement changes alongside performance changes.

This is particularly important for B2B businesses with long sales cycles. A paid campaign can generate a lead today, an MQL weeks later and an opportunity months later. If the measurement architecture changes in the middle of that journey, a simple month-on-month ROAS comparison can become misleading.

The New Definition of Reliable Marketing Measurement

Reliable measurement does not mean every user is perfectly observable. That is no longer a realistic standard.

Reliable measurement means you understand what is observable, what is modelled, what is missing, why it is missing and how those limitations affect business decisions.

For B2B marketers, the goal should be a measurement system where:

• Consent is respected.

• Tags behave predictably.

• Conversions are not duplicated.

• Regional differences are documented.

• Platform data is reconciled with CRM data.

• Modelled results are clearly distinguished from observed results.

• Management understands the confidence and limitations behind the numbers.

Final Takeaway

Consent Mode and Google Signals are not simply configuration settings inside Google Analytics. They are part of a broader shift in how digital marketing measurement works.

The organisations that adapt best will not be the ones with the most dashboards. They will be the ones that understand their measurement architecture well enough to know when the numbers are reliable, when they are modelled and when further validation is required.

For performance marketing teams, that changes the question from “What did Google report?” to a much more useful one:

“What can we confidently say happened — and what part of the journey is still invisible?”

Editorial Note

This article focuses on marketing measurement and implementation considerations, not legal advice. Privacy and consent requirements can change and should be validated with qualified legal/privacy specialists for the relevant markets.

Privacy Policy

Terms & Conditions