The YouTube universe, scored, ranked and ready to plan with.
A continuously refreshed graph of every channel that matters in a market, a suitability rating per channel, and a terminal to plan from. That part is live on real plans. ROMI scoring and cross channel come next.
Three layers, and every number traces back down.
Every figure on the terminal traces back to a specific channel, a model we can explain, and the constraint it was scored against.
Decision
Reallocation, scenario planning, audit-grade reporting.
Intelligence
Fit and safety scores, incrementality, saturation.
Data
Every channel as a node. Normalised, queryable, refreshed.
Every score at the top traces back to raw channel data at the bottom.
A normalised, queryable graph.
Every channel is a node carrying what it is, who watches, whether it is safe and whether it returns. Edges track overlap and substitution between channels. Inventory refreshes daily, safety weekly, ROMI calibration monthly.
Identification
Canonical channel and video IDs across markets
Audience scale
Subs, reach, country distribution
Engagement
Retention curves, completion, comment velocity
Brand safety
Industry frameworks, verification signals and historical flags, composited
Vertical classification
Embedding based taxonomy, multi label
Audience demographics
Age, gender, geo, interest clusters
Commercial signals
Sponsorship density, CPM bands, terms
Attention metrics
Quality weighted view time
Incrementality signals
Holdouts, geo lift, MMM contributions (coming soon)
Behavioural signals
Time of day, device and context patterns, panel calibrated
Brand safety scored like credit risk.
A composite rating per channel, from industry frameworks, verification signals, platform signals, our own incident graph and analyst review. Then the part most vendors skip: safe inventory ranked by incremental business contribution.
- CompositeContent suitability, context risk, audience exposure, historical incidents, monetisation stability
- RatingA+ to D, with a confidence score per channel
- KidsFour layer child audience exclusion, our own classifier in the lead. Part of the first delivery.
- ThenSafe media is not necessarily effective media. Buy, review, avoid, exclude.
Next up: what each channel actually returns.
ROMI per channel, calibrated with incrementality holdouts, and a planner that lets you move budget on screen. Decision intelligence, not autonomous trading. Humans decide. The graph compounds.
Observe
Continuously read market state from the graph.
Detect
Flag the saturation, drift and waste as it shows up.
Simulate
Score reallocation scenarios under constraints.
Recommend
Surface ranked, explainable recommendations.
Decide
The buyer or planner approves. Architect never trades silently.
Calibrate
Outcomes feed back into the graph.
The terminal is what you see. The graph underneath is what makes the ratings worth trusting.
What is live and what is next.
What is live runs on real plans today. What is next is in active development.
Channel intelligence for YouTube
A suitability rating per channel, A+ to D with a confidence score, a filterable terminal of the brand safe universe for a market, and a waste finder that names the channels that pass every safety check and still do nothing.
Kids audience exclusion
Four layer child audience screening with our own content classifier in the lead.
ROMI scoring per channel
Return on media investment scored per channel, calibrated with incrementality holdouts.
Cross channel graph
Social, CTV, programmatic and OOH on the same ratings scale. One plan, one graph.