When Ad Blocking Becomes a Measurement Problem

By James Ensor, Global Lead, Server-Side, Didomi

Ad blockers are designed to stop unwanted advertising from reaching the browser. That is their obvious function. Their less visible effect can be far more significant for marketers.

The same technology that prevents an advert from loading can also block the requests used to measure what happens on a website. A visitor may still arrive, browse and convert, but the analytics system may never receive the information needed to record those actions.

That creates a particular problem for marketing teams. When ad blockers remove parts of the measurement trail, performance data can become incomplete without necessarily looking wrong. The traffic has not disappeared, but it has become harder to see.

The issue is especially relevant because ad blocker usage is highest among younger adults, higher earners, college-educated people and employed people with children. These are the very audiences that marketers are often trying to understand and reach.

Ad blocking can remove more than advertising

The way ad blockers operate helps explain why the impact reaches beyond advertising.

Browser extensions inspect requests made as a page loads, comparing them against regularly updated blocklists containing domains, scripts and other technical signatures. When a request matches one of those signatures, it can be stopped before it leaves the browser.

That can include familiar measurement infrastructure. Requests to domains such as `googletagmanager.com` can be blocked, while scripts associated with Google Tag Manager or Google Analytics may never get the opportunity to record an event. In other words, measurement can fail before there is anything for an analytics platform to report.

There is an important distinction around consent here. A visitor can accept a site’s cookie banner and agree to measurement, yet the browser can still prevent the resulting request from reaching its destination. Consent determines whether data may be collected. It does not guarantee that the technical collection will succeed.

That works both ways. Finding a way to recover measurement for consenting visitors does not change the choice of someone who has declined tracking. Those users should remain untracked. The objective is to address the technical gap affecting people who have already agreed to measurement.

When missing signals start changing the picture

Once enough activity goes unrecorded, the consequences can move quickly from analytics into commercial decision-making.

If sessions, conversions and channel-attributed revenue are under-reported, campaigns can appear to be performing less effectively than they really are. Attribution becomes less reliable, while cost per acquisition can appear higher and return on ad spend weaker. That can influence how budgets are allocated, including decisions to reduce investment in channels whose results have simply been under-measured.

The same missing signals affect audience activation. A visitor who is never recorded cannot subsequently enter a remarketing audience. Smaller audience pools also provide less information for creating lookalike audiences, while advertising platforms have fewer conversion signals available for optimisation.

What makes the problem particularly difficult is that the resulting data can still look credible. There is no obvious warning that says a proportion of visitors was never measured. A dashboard can continue to produce clean reports while describing only the activity that made it through the collection process.

For marketers, that makes ad blocking a measurement issue as much as an advertising one.

Moving measurement out of the browser

The architecture used to collect data therefore matters.

With client-side tracking, scripts run in the browser and requests are made from there. That puts the measurement process directly in the environment where ad blockers can inspect and interrupt it.

Server-side tracking changes where that process takes place. Data can be sent to infrastructure controlled by the organisation, using its own domain as a subdomain or reverse path, before tags fire from that environment. This creates a first-party context rather than relying entirely on third-party scripts that blocklists may already recognise.

There are other benefits to this approach. Fewer third-party scripts need to be loaded in the browser, which can help reduce the burden on the page. Organisations also gain greater control over the data passing through their infrastructure, including what information is subsequently shared with individual partners. Governance can therefore be built into the collection architecture itself.

Server-side tagging does not automatically solve the problem, however. If the browser’s initial request still follows a recognisable pattern, a sophisticated ad blocker can intercept it. The implementation therefore needs to account for how tracking components are served, including using URLs that do not match known blocklist signatures.

Recovering the data is only the first step

The practical impact can be significant. In a case study involving a subscription publisher serving healthcare professionals, moving the setup server-side and hosting its tag manager in first-party mode under a renamed file helped prevent ad blockers from interfering with paywall gating. The publisher reported a 10% increase in paywall impressions and a 10% increase in page views.

The wider lesson is that measurement starts well before a marketer opens a dashboard. If the relevant signal cannot make it through the collection process, every subsequent analysis is working with less information.

Recovering that signal is only part of the task. Once data is being captured again, it still needs to reach the tools and decisions it was collected to support. Otherwise, organisations risk creating data that is technically available but effectively disconnected from the marketing activity it was intended to inform.

A resilient approach to measurement therefore needs to account for the entire journey of consented data. Ad blockers make the weaknesses in that journey particularly visible, but the underlying principle is broader: marketers need confidence that the information they are using to understand performance represents the activity they are actually trying to measure.