Contextual targeting is not the fallback. It is the reset.

Last week at Programmatic Pioneers 2026, Skyrise's Jonny Whitehead moderated a debate on contextual targeting without cookies. The conversation could have stayed in familiar territory: signal loss, identity, measurement, regulation, replacement technology. It did not.

The strongest point was more uncomfortable. The end of cookies is not a targeting problem. It is a quality problem.

For years, too much of digital media has been built around the promise of individual-level precision. Find the user, follow the user, attribute the user. That logic is now under pressure from every direction.

Regulators are looking past cookies to other forms of storage and tracking. Advertisers are questioning what digital attribution really proves. Buyers are asking whether cheap reach is worth much when the impression is low quality, poorly measured, or not even human.

The answer is not to rebuild the same system with different plumbing. It is to ask better questions. Where did the ad appear? What was the person doing in that moment? Was the environment trusted? Was the creative right for that context? Did the channel do the job it was asked to do? That is the point where contextual targeting stops being a privacy-safe alternative and starts being a reset.

Replacement is not the same as progress

One of the clearest points from the debate was that "cookieless" does not automatically mean better. Much of the market has treated the loss of cookies as a technical migration: swap one identifier for another, move from one storage mechanism to the next, keep the attribution models alive by other means.

That misses the bigger shift. The real question is not whether a cookie is present. It is whether the underlying approach reduces privacy risk? Does it improve media quality, and does it give advertisers evidence they can trust?

If a new signal preserves the same behaviours, it is not progress. It is substitution. The more interesting opportunity is to leave the old race to the bottom behind. The one that was built on cheap impressions, cheap reach, weak context, and questionable value. Context offers a way out. Not because it is simpler, but because it is richer.

Context is no longer a keyword list

Contextual targeting is still too often judged by an outdated version of itself. For some buyers, context means a keyword list, a broad category, a blunt brand-safety filter. That is not where the market is heading. The better version reads the full environment around an impression. It understands attention, quality, relevance, sentiment, format, location, time, audience concentration, and creative fit.

A football article is not simply "football". It can carry excitement, rivalry, loyalty, or disappointment, and the mood around it shifts with the result. A connected TV impression is not simply big-screen video. It sits inside a household, a piece of content, and a supply path. A programmatic out-of-home screen is not simply a frame. It has a location, a surrounding environment, a time of day, a movement pattern, and an audience profile.

The point is not to make context sound complicated. It is to make it more useful. The strongest contextual strategies do not only decide where an ad can appear. They help decide what should be said, how it should be said, and what value that impression should carry.

Creative is more important than ever

One of the sharpest lines from the discussion was that creative is becoming the new targeting.

As identifier-led targeting becomes less dependable, creative has to carry more of the load. Not as decoration applied after the media plan is built, but as a core variable in how performance is produced.

The debate pointed to Meta's Andromeda as one marker of this direction: platforms using large volumes of creative variation to learn which hooks, messages, and assets work for different audiences. The broader point holds well beyond any single platform. When the system has less confidence in who someone is, the ad itself has to work harder in the moment.

That means more than producing more assets. It means connecting creative to context. The right message in the wrong environment is still waste. The right environment with generic creative is still underpowered.

Contextual targeting only reaches its full value when the creative is built to respond to the moment it appears in. So the planning question changes. Not only who are we trying to reach, but what are they reading, watching, passing, or feeling when we reach them, and what creative earns that moment.

New channels should not repeat old mistakes

The same question ran through the day's discussion of connected TV and programmatic out-of-home. Different channels, different risks, one shared lesson: transparency cannot be retrofitted after scale.

The open web has already shown what happens when growth outruns governance. Supply paths turn opaque, measurement fragments, fraud adapts, and trust becomes hard to rebuild. Connected TV and programmatic out-of-home are still forming their programmatic habits, which makes this the moment to get the basics right.

The fundamental questions are straightforward: identify your sellers, track the investment, and define what truly constitutes a viewed impression. Ensure delivery is verified, understand where fraud might manifest, and define "premium" based on your specific objectives rather than marketing materials.

Connected TV brings scale, premium environments, and high attention, alongside fragmentation and fraud risk. Programmatic out-of-home carries the advantage of physical assets and far lower fraud risk, but a real screen does not settle every trading and measurement question. The most useful phrase of the day was also the simplest: always-on audit. That should be the standard. Not because buyers distrust every partner, but because trust is built by checking.

From precision theatre to useful evidence

Digital advertising has spent years selling certainty, and that certainty is wearing thin. Individual-level attribution was never as clean as the dashboard made it look. The rise of artificial intelligence agents, non-human traffic, and more complex privacy rules only makes that clearer.

The future is not measurement without imperfection. It is measurement with better judgement. Different channels need different evidence. Attention may matter in one context, incremental reach in another, and footfall, brand lift, dwell time, creative engagement, or business outcomes elsewhere. The goal is not to force every channel into the same reporting shape. It is to understand whether the channel did the job it was chosen to do.

That sounds obvious. It is still not how enough media is planned. Too often the market rewards what is easiest to count rather than what is most likely to work. Contextual targeting asks for a more grown-up model: less precision theatre, more useful evidence. Less chasing the individual, more understanding the moment. Less cheap reach, more quality, relevance, and accountability.

That is why contextual targeting should not be treated as the fallback after cookies. It is the chance to build a better system. One that respects privacy. One that values environments. One that gives creative a bigger role. One that holds new channels to a higher standard before bad habits set in.

Cookies made digital advertising feel precise. Context can make it better.