Open Data Infrastructure
Answer six honest questions about how your data actually moves, who can read it, and whether an AI agent could safely query it today. Two minutes, no email required — you get a real score and the specific things to fix first.
The readiness question is simple: if the business needs to use its data across tools, teams, clouds, and AI systems, how much friction does the infrastructure create? Score yourself the way you would in an architecture review — on what is true, not what is on the roadmap.
Keep going
The scorecard tells you where you stand. These pillars tell you what to do about it — written for technical leaders, no vendor pitch.
Six dimensions decide whether your data is an asset you control or a liability you rent. Each is scored 0–3. The interactive tool above uses exactly this rubric — it is written out here so the framework is useful even without it.
Strategic (3): critical operational and analytical data is reachable through durable APIs, exports, event streams, or change data capture you could repoint.
Warning sign (0–1): access depends on manual extracts, screenshots, undocumented endpoints, or vendor services that make portability expensive.
Strategic (3): data is stored in open formats with metadata that multiple engines can read and respect.
Warning sign (0–1): useful data is trapped behind proprietary execution, hidden storage, or a single vendor's metadata system.
Strategic (3): teams discover tables, owners, schemas, freshness, lineage, permissions, and quality signals from a shared catalog layer.
Warning sign (0–1): tribal knowledge is the primary metadata system.
Strategic (3): the platform supports multiple engines and workloads without forcing unnecessary data copies.
Warning sign (0–1): each new workload requires a separate data mart, export process, or vendor-specific integration.
Strategic (3): policies, access controls, audit trails, data quality, and lineage are designed into the platform.
Warning sign (0–1): governance happens after the data has already spread to downstream systems.
Strategic (3): agents and AI apps retrieve governed data and context with clear permissions, provenance, and fallback behavior.
Warning sign (0–1): AI demos work only with hand-curated extracts or over-permissioned access.
| Score | Meaning | Practical implication |
|---|---|---|
| 0 | Not present | The capability depends on manual work or does not exist. |
| 1 | Ad hoc | The capability exists in isolated tools or teams but is not reliable infrastructure. |
| 2 | Operational | The capability is dependable for core workflows but not broadly interoperable. |
| 3 | Strategic | The capability is open, governed, reusable, and ready for AI or cross-platform use cases. |
Interpretation: a high score does not mean every tool is open source. It means your data, metadata, and governance model are open enough that the organization — not a vendor — keeps strategic control. The lowest two dimensions are where an AI initiative will stall first, regardless of model quality.