Narrativ is a data company built around a simple architecture:
We collect two categories of inputs:
Managed Sources
External APIs and web scrapers that produce structured public and system-level data. These are operated and maintained by Narrativ.
User-Sourced Data (Permissioned)
Users connect accounts and choose what they approve. This can include both public signals and private signals where consent is explicit.User-sourced data is always opt-in. No data is collected without explicit approval. This is what makes Narrativ's downstream products compliant by design.
Processing (how value is created)#
Those inputs flow into a proprietary processing layer that turns raw activity into:Signals — normalized events, cleaned, timestamped, and tagged so they can be compared across sources and users.
Indices — computed metrics, aggregated and weighted measures derived from one or more signal streams. Indices are the core unit that powers everything downstream.
Outputs (what indices power)#
Indices power two product surfaces:B2B data products — high-signal, compliant datasets and time series sold to institutional buyers.
Time-boxed markets — markets that resolve on index outcomes, a market layer wrapped around the data engine.
Rewards (how value flows back)#
Value flows back to users through recurring prize pools funded by a portion of platform fees and downstream data revenue. Access to prize pools is gated by opt-in data sharing, aligning contribution with upside.The reward loop closes the flywheel: more data shared → better indices → more revenue → larger prize pools → more users willing to share.
