Home · Screening API · On-premise deployment

Screen inside your own network

For teams that cannot send customer names outside their walls. Load our collection files into your own environment, screen there, and fetch only new releases from us. Everyone else can use the hosted screening API.

Status. The bulk-file route below is available today. A packaged, self-hosted matching server is not yet published; its packaging and requirements will be listed here when it is.

How it fits together

Your systems screen against a copy of the data held on your own infrastructure. The only traffic to us is your copy fetching new releases.

Self-hosted screening Crown Intelligence publishes collection files. Inside your network, a loader fetches new releases into a local screening service, which your onboarding, payments and case systems query. Customer data stays inside your network. Crown Intelligence collection files + index Your network Loader checks index.json Screening service local copy of the data Onboarding Payments Case review
Names you screen are queried locally; only published files cross the boundary.

Why run it yourself

  • Customer data stays with you. The names you screen never leave your environment, which can simplify your obligations under PIPEDA, Quebec's Law 25 and provincial privacy law, and your own third-party risk review.
  • The data licence is what you pay for. A screening licence covers the data and its updates.
  • Your own throughput. Speed depends on your own infrastructure, not on a hosted rate limit.
  • Your own lists alongside ours. With the bulk files in your own screening engine, internal watchlists can sit in the same screening step.
To be provided: the licence terms for the self-hosted server software itself (if shipped), whether a packaged server would support loading a customer's own watchlists, and whether a flat-rate licence removes all volume limits.

How to get started

  1. Decide on hosted or self-hosted

    Weigh privacy, latency and the effort of running a service. The questions to prepare cover the trade-offs.

  2. Try matching against the hosted API

    Send a few of your own test records to /api/match to see the scoring and the fields you will work with.

  3. Agree a data licence

    Self-hosting needs a screening licence, or a reseller licence if the service runs inside a product you sell. Talk to sales.

  4. Load the files and keep them current

    Fetch the collection files, load them, then apply updates as described below. The bulk-file guide documents the layout.

Server requirements

For the bulk-file route, the requirement is whatever your own screening engine needs to hold the collections you license; file sizes are listed on each download.

To be provided: for a packaged self-hosted server, how it ships (for example containers and an orchestration file) and measured minimum and recommended memory, CPU and disk.

How updates arrive

Each collection publishes an index.json that says which version is current and when it was built, with full files and a delta.json of changes. A loader checks the index on a schedule and applies the delta, or reloads in full.

Collection index

Sources refresh on their own publishers' schedules, so how often a new release appears depends on the collection. See freshness.

Support for self-hosted users

We cannot see inside your environment, so we cannot monitor your copy. For questions about the files or the data, get support; to plan a deployment, talk to sales.

To be provided: whether deployment consultation is offered, and on what terms or fees.