Consent or Pay: How 'Accept Tracking or Subscribe' Models Work
The binary banner, accept or reject, assumed the service was funded either way. Consent or pay breaks that assumption openly: the publisher tells the visitor the content is paid for by advertising built on tracking, and offers the alternative of paying directly. Hundreds of media sites now run some version of it, and the model has become one of the live questions in European privacy regulation.
This article explains the mechanics, the regulatory arguments on both sides, and the design details that separate a defensible implementation from a paywall wearing a consent banner as a costume. Whatever your view of the model, it is spreading beyond news into apps and services, so the analysis travels.
The mechanics of the model
A visitor declining tracking is offered a subscription, usually a modest monthly fee, for access without the advertising trackers. Accepting remains free. The implementation needs care: the paid tier must actually suppress the ad stack, not merely hide the ads, and the consent and subscription states must reconcile across devices through the account. Publishers adopted the model because the reject all era cut their addressable advertising sharply; the pay option recovers a fraction of decliners as subscribers and, frankly, increases acceptance among the rest, which is precisely the effect regulators are examining.
The regulatory argument, both directions
Supporters argue the model makes an honest trade explicit: content costs money, and the visitor may pay in attention and data or in cash, with a real choice either way. Several national regulators have accepted versions of it with conditions. Critics answer that consent extracted under a paywall is not freely given, that the fee functions as a penalty on a legal right, and that the model prices privacy as a luxury. European level guidance has pushed toward requiring a genuinely equivalent alternative and scrutinizing fees, with particular attention to large platforms. The practical reading: the model is neither clearly lawful everywhere nor clearly banned; it is conditional, and the conditions are tightening.
What makes an implementation defensible
Four properties recur in the guidance. The fee should bear a reasonable relationship to the advertising value per user, not a punitive multiple of it. The paid experience must be genuinely tracking free in the advertising sense, verified by audit rather than asserted. The choice architecture must be neutral: equal visual weight, no pre selection, no loss framing against the paid path. And rejecting without paying should remain possible for at least a limited form of access in some regulators' views, such as a version with contextual, non tracked advertising. A publisher who cannot demonstrate all four is running a steering pattern with a price attached.
Lessons for sites not using the model
Even if you never charge, consent or pay clarifies the general principle regulators are converging on: the alternatives to consent must be real. A reject path that degrades the service beyond what the missing data justifies draws the same criticism as an unreasonable fee. Audit your own reject experience with that lens. Does declining remove features that never depended on tracking? Does the site nag decliners on every visit while remembering accepters forever? The distance between your reject path and your accept path is the number the whole debate is about, whatever currency it is measured in.
Related reading
These earlier guides cover connected questions in more depth: Consent and A/B Testing: Your Experiments Are Tracking Too, The Accessible Consent Banner: Screen Readers, Keyboard Paths, and Choices That Exclude Nobody, and Web Workers and Consent: The Scripts Your Wrapper Cannot See.
Frequently asked questions
Is consent or pay legal?
It depends on the jurisdiction and the implementation, and the position is evolving. Some national regulators have accepted conditioned versions; European level opinions are more skeptical, especially for dominant platforms. Any publisher adopting it needs current, local legal advice rather than a general answer.
What is a reasonable fee for the paid tier?
Regulators look for a relationship to the per user advertising revenue the tracking would have generated, which for most publishers is a small monthly figure. Fees that read as deterrents rather than equivalents are the ones drawing findings.
Does the paid tier have to remove all cookies?
No. Strictly necessary and functional storage remains. What must go is the advertising and profiling stack the visitor declined. Publishers should verify with the same tag auditing used for any consent state, because ad partners have a way of persisting.
Could this model come to ordinary ecommerce or SaaS?
Elements of it already appear wherever a free tier is data funded. The scrutiny follows the power imbalance: the more essential the service and the larger the provider, the harder regulators look at charging for the private option.