Rise Launches Agentic Bid Enrichment: Real-Time Intelligence on Every Bid Request
by on 1st Jun 2026 in News
Rise, the intelligent supply platform, today (1st June, 2026) announced the launch of Agentic Bid Enrichment, an AI-powered intelligence layer embedded directly inside its platform that generates real-time intent signals on bid requests, including traffic with no user identifiers.
According to the IAB's State of Data Report, 50-60% of programmatic inventory carries no user ID, and in Rise's own analysis, roughly 50% of bid requests carry misclassified or missing IAB category labels, leaving demand either blind to the opportunity or targeting the wrong audience.
The result is a gap between the scale of available inventory and the actual understanding of each opportunity. Agentic Bid Enrichment resolves it at the source.
For every bid request, Rise's agentic layer synthesises three signal types in real time: what the page is about right now, the user's environmental context, and learned behavioural data where available. These are not treated as independent inputs. They are synthesised into a single, coherent understanding of user intent and injected directly into the OpenRTB bid request alongside a proprietary Confidence Score.
The Confidence Score is generated and owned by Rise. It reflects how strongly intent is expressed and enables DSPs to calibrate bid strategies with higher precision.
Beyond intent classification, the agentic layer can identify whether the user behind a request matches known conversion patterns, surfacing that signal in real time, inside the auction.
All of this happens in milliseconds, at full programmatic scale, natively inside the supply path.
"The real challenge is not just building a model that reasons well but running it across billions of bid requests a day and having the economics make sense. Developing and training agents under those constraints is a significant engineering undertaking. Our architects have invested heavily in a hybrid inference architecture designed specifically for this. We have been building toward this for years, investing in infrastructure, architecture, data pipelines, workflows, and strong partnerships that hold it all together, and that groundwork is what makes the difference." - Noa Keydar, VP data, Rise.
In early results from live traffic, the agentic layer consistently generated intent classifications on requests that previously carried no usable targeting data. IAB category coverage increased from 50% to 100%. The average Confidence Score across enriched requests was 87%. In cases where categories were present, the misclassification rate was 30%, corrected to zero after enrichment. CPMs on unclassified and generic inventory lifted by up to 3x once intent was established.
Rise offers implementations ranging from fully managed supply optimisation, where Rise acts as a smart SPO layer with no integration required on the buyer side, to direct signal sharing for DSPs that want to apply enriched data inside their own systems, through to agent-to-agent integration aligned with emerging ADCP standards.
Follow ExchangeWire