Diaries Are Dead. Truth Lives Here.
Audience360 is a proprietary passive cross-media measurement currency. Built on on-device audio fingerprinting, it captures what people actually heard and saw — not what they remember hearing and seeing. The difference between recall and reality is the difference between guessing and knowing.
POPIA compliant by architecture, not by afterthought. Privacy is not a feature we added — it is a constraint we designed around from day one.
The Audience360 app runs as a foreground service on participant devices. When any media plays nearby — TV, radio, podcast, streaming — the app captures a brief audio sample. This sample is immediately converted into a non-reversible acoustic hash (fingerprint) on the device itself. The raw audio is discarded instantly. It never leaves the phone.
Only the hash and a coarse location cell are transmitted to our matching servers. There, the hash is compared against our pre-indexed library of broadcast and streaming content. A match tells us exactly what the person was exposed to, when, and roughly where.
No raw audio ever leaves the device. No conversation is recorded. No private data is transmitted. The hash is non-reversible — you cannot reconstruct audio from it, even with full access to the algorithm.
The technology is fundamentally different from a diary or recall survey — it captures what actually happened, not what someone remembers. There is no recall bias, no social desirability bias, no fatigue effect. Just ground truth.
Audience360 is engineered for massive concurrent device panels. Hundreds of monitored broadcast and streaming channels are fingerprinted simultaneously. Millions of fingerprint matches are processed per hour with near real-time latency.
The system scales horizontally. Adding more panellists, more channels or more markets does not degrade performance. The architecture was built for continental scale from the beginning.