Ad blockers: circumventing the user is the wrong answer

A substantial part of your audience is invisible to your measurement tools. Nearly a third of internet users run a blocker, and since the shift to explicit consent, many others simply refuse to be tracked. Your dashboards therefore reflect only a fraction of reality — and that fraction is biased. The instinct, faced with this, is to recover the missing users by circumventing whatever makes them invisible. That is a mistake, and on three counts at once. The right question is not “how do we see everyone”, but “how do we measure what matters without tracking anyone”.

The blind spot is real — and it is biased

The figures leave little room for doubt. Around 31.5% of internet users aged 16 to 64 say they use a blocker at least occasionally, on the order of 912 million people (GWI data, 2025). In Europe the proportion climbs above 40%, and exceeds 49% in Germany. To this is added, since explicit consent became the norm, the far from negligible share of visitors who actively decline tracking. A real fraction of your audience therefore never appears in your statistics.

The problem is not merely one of missing volume. It is qualitative. The users who block or decline tracking are not a representative sample drawn at random: they are, proportionally, the most technical, the most informed, the most privacy-conscious profiles. In other words, precisely the segments an organisation often has the greatest interest in understanding. Your dashboards do not merely undercount; they over-represent one population — the least equipped, the least wary — at the expense of another. A decision optimised on this data is optimised for the wrong audience.

This effect is even more pronounced in the world of business applications. The users of a professional tool are often recruited from those same technical, demanding profiles. The invisible segment is not a margin: it is sometimes the very core of usage.

Why circumventing the user is the wrong answer

Faced with this blind spot, one response has circulated for a long time: detect the blockers and find a way to track their users anyway. This is the approach illustrated by the tutorials of a bygone era — that of Universal Analytics, now replaced by GA4. This response fails on three fronts.

Technically, it is an arms race lost in advance. Blockers update faster than the workarounds, and anti-blocking devices degrade the experience — warning messages, locked content, features that break at the next update cycle. You invest in permanent maintenance for a battle you lose again with every iteration. The evolution of browsers, which are progressively tightening their extension rules and abandoning third-party cookies, only accelerates this obsolescence.

Legally, the exposure is direct. Tracking a user who has explicitly signalled their refusal — through a blocker or a declined consent — runs head-on into the foundation of the GDPR and the ePrivacy directive, which rests on consent. For an application that handles professional data, this is a risk the IT department carries, and one that is not justified by a gain in measurement.

On trust, finally — and this is the most lasting. Surveilling people who have said no erodes the most precious capital of a relationship, especially in B2B: trust. Winning the tracking battle only to lose the reputation war is a poor calculation. The blind spot is real, but the cure of circumvention is worse than the disease.

Measure differently: move the point of measurement

The solid path does not consist of tracking better, but of moving the point where you measure. Historical tracking rests on a script that runs in the user’s browser — exactly where the blocker acts and where consent applies. The whole challenge is to break free of that dependency.

Server-side measurement is the first lever. Rather than depending on a third-party script that the browser can neutralise, you record the events that matter — a conversion, a key action, an API call — where they actually happen: on your own infrastructure. Whether or not the browser runs a tracker, the business event itself has indeed taken place, and you have a record of it. Conducted transparently and with respect for consent, this approach is both more robust against blockers and cleaner where data is concerned.

Consent Mode illustrates a second logic, that of modelling. Since March 2024, Google’s Consent Mode v2 — made mandatory for users in the European Economic Area under the Digital Markets Act — embodies a change of paradigm: when a user declines, you do not track them; you estimate the gap from consented behaviour. You accept not seeing everything, and you model rather than force collection. It is the exact opposite of circumvention.

Privacy-respecting analytics tools extend this idea. Designed without individual tracking — no cookie, no fingerprint — they measure the audience in aggregate. They are by nature less hampered by blockers, since there is nothing to block on the tracking side, and natively more aligned with regulatory requirements. The debate between these solutions and conventional advertising tools deserves to be had upstream of a project, not endured after the fact.

The deepest shift, however, is conceptual: measuring outcomes rather than individual journeys. Most of what an organisation seeks to know — how many conversions, which pages perform, which channel brings in customers — can be read in aggregate signals and server-side events that the blocker does not touch. You do not need to follow each individual to know what works.

What a CIO must weigh

Let us be honest about the trade-off: privacy-respecting measurement offers less individual granularity. You trade some precision on the user journey for legal soundness, durability and trust. For an organisation that lives on its reputation and handles professional data, this is almost always the right trade.

But it is not a matter of giving up measurement. Well designed — combining server-side measurement, respect for consent and aggregation — a measurement strategy provides reliable, durable signals, sufficient to decide. The illusion to abandon is that of perfect individual tracking, which no longer exists in 2026 in any case, caught between blockers, the consent framework and the evolution of browsers. You are not losing something you had; you are adapting to what is real.

That is where the decision lies. Measurement is a choice of architecture, not a tool setting bolted on at the end, nor a battle delegated to an anti-blocking module. What do we really measure? Where do we measure it, browser-side or server-side? Under what consent regime? These questions are settled upstream, at the design stage, and it is the role of the IT leadership to pose them.

Measure what matters, not everyone

The blind spot of blockers is not a defect to be patched by outrunning the user. It is a signal: individual surveillance on the browser side is a declining foundation, caught between blockers, the law of consent and the evolution of browsers. The robust path — server-side measurement, respect for consent, aggregation — happens to be the compliant path and the durable path as well. This convergence is no accident.

The role of an IT leadership is not to see every user — that was never possible, and is becoming less and less legal. It is to measure what is needed to decide, soundly and lastingly. The right answer to the blind spot is not a more powerful telescope trained on people who have drawn the curtains. It is to measure what matters, and to accept not counting everyone.