The next wave of curation and media quality
Opinion
As AI models advance and more attention and semantic signals enter the mix, curation will continue to shape programmatic strategy. Onetag’s global marketing director explains.
Programmatic buying has spent years trying to climb out from under its own weight. What began as a way to organise inventory and automate transactions gradually buried buyers in an expanding flow of impressions that looked efficient on paper but rarely behaved that way in practice.
Sell-side curation first emerged as a remedy for that clutter. It offered a way to streamline the supply and give campaigns a clearer path into high-quality environments. The role it plays today is changing. The industry is beginning to view curation, with its real-time sell-side decisioning, as foundational, shaping the connection between context, attention, and outcomes.
As digital advertising leans deeper into AI, the pressure to understand quality has increased. More signals exist, increasing complexity and fragmentation. Advertisers are seeking paths that deliver real impact, and many of the usual shortcuts no longer work.
The most interesting progress in solving these challenges is happening where the curation of performance data, semantic insight and audience behaviour come together.
Buyers who build their strategies on these signals often find that quality becomes easier to define and easier to apply. They also come to understand the value of their first-party intelligence more clearly, as curated environments tend to make those data assets more actionable.
This shift is also helping restore confidence in the open web. There was a time when open programmatic felt too unpredictable, too crowded or too inconsistent to support ambitious brand goals.
The picture is changing because supply can now be shaped earlier and with greater accuracy.
That change is encouraging for publishers as well. When their strongest environments surface more consistently, their investment in meaningful content yields more reliable returns. It also counters the pull toward the closed ecosystems of big tech, which often promise control but restrict how data can work across the wider market.
Understanding content in a more precise way
The foundation of this evolution lies in how content is interpreted. Traditional contextual approaches leaned heavily on keywords or broad categories. They were often blunt and struggled to convey a page’s nuance or intent.
Semantic analysis brings a different depth. It identifies concepts, themes, entities, and relationships that reflect genuine meaning.
It gives buyers a clearer view of where their messages are landing. It also widens the field, revealing quality environments that may never have been surfaced by lists or manual selection.
When campaigns rely on semantic understanding, they tend to perform with more consistency. They also become more relevant to the user’s moment, which can support stronger engagement.
The value of this is easier to see when it is connected to attention metrics. Attention is no longer limited to a single measure. It can reflect time-in-view, the balance between content and advertising on a page, the likelihood that an ad is positioned where users naturally focus, and other signals that describe the tone of the environment.
These two areas work well together. Semantic insight explains the context. Attention signals explain the experience. When both are fed into optimisation modelling, campaigns learn quickly and adjust without being pulled off course. This blend also reduces the risk of decisions being made in isolation, a challenge for organisations navigating growing fragmentation across their technology stacks.
The role of continuous feedback
A defining characteristic of modern curation is its ability to respond in real time.
Buyers once had to wait for reporting cycles or mid-flight analysis to understand whether supply paths were working. Today, things move more fluidly.
If an environment repeatedly falls below the attention, quality, or performance thresholds that a buyer has set, it can be filtered out instantly. If a content category starts producing stronger-than-expected results, the system can elevate its better-performing ad placements across the premium open web.
These decisions support a continuous learning loop. Each impression carries information, and that information shapes the next choice.
This cycle is becoming essential as signal loss accelerates. Without third-party cookies or reliable identifiers, a large part of campaign effectiveness depends on how well environments align with intent and behaviour. Real-time feedback helps bridge the gap. Buyers see more predictable results, and publishers see improved recognition of their strongest inventory.
A clearer path for publishers
Publishers have a significant stake in how curation evolves. When quality is defined more rigorously and impression decisioning occurs earlier in the supply chain, strong content can shine through more consistently.
Premium placements that once risked being overlooked now have a clearer path to demand. That creates healthier economics for the open web. Many publishers are also beginning to view curation as a way to support their long-term strategies. It rewards environments that cultivate authentic engagement, and it recognises the editorial investment behind them.
There is also a practical benefit. When curated demand shows higher bid density, more predictable pricing, or steadier revenue, publishers can plan with greater confidence. That stability is important as they navigate new formats, audience shifts, and ongoing pressure from walled gardens.
What comes next
Curation is moving into a phase defined by adaptability. Buyers want supply paths that adjust to outcomes rather than static configurations. Publishers want a clearer understanding of how their quality is measured. Both sides want increased effectiveness, minimised complexity and simpler workflows.
As AI models advance and more attention and semantic signals enter the mix, curation will continue to shape programmatic strategy.
The industry is moving toward a future in which media quality is no longer an abstract concept. It has structure. It has a measurement. It has a direct line to performance data.
Curation with sell-side decisioning intelligence is becoming the framework that connects these pieces, guiding investment toward environments that truly deliver and strengthening the foundation of the open web in the process.
Keith Arrowsmith is the global marketing director at Onetag