AI Readiness Assessment

Scattered across spreadsheets and systems, inconsistent formats, no ownership Centralized in a few systems but with known quality issues and gaps Mostly clean and centralized, with occasional manual cleanup needed High-quality, well-governed data with documented lineage and quality checks

Locked in silos; requesting access takes weeks or requires IT tickets Accessible to a few technical staff via direct database access Available through APIs or a data warehouse to most technical teams Self-serve access with proper permissions for both technical and business users

No, data lives only in operational/transactional systems A basic reporting database exists but is not maintained rigorously Yes, a warehouse/lake exists and is used regularly by analytics teams Yes, a modern, well-maintained platform serves as the source of truth company-wide

Largely informal; no documented policies Basic policies exist but are inconsistently enforced Documented policies are enforced with periodic audits Mature, audited practices with automated enforcement and monitoring

No policy; employees use whatever tools they like without guidance Informal guidance exists but is not documented or enforced A written policy exists and is communicated to staff A living policy exists, is enforced, and is reviewed as tools evolve

We do not have a process for this today Risk is considered informally, on a case-by-case basis A defined review process exists for higher-risk use cases A formal risk framework applies consistently across all AI initiatives

No one; AI efforts are ad hoc and uncoordinated An individual owns it informally alongside other responsibilities A dedicated owner or working group leads AI strategy Executive sponsorship exists with a cross-functional governance committee

We have not assessed our exposure to AI regulation We are aware of relevant regulations but have not taken action We have assessed exposure and begun addressing key gaps We actively track and comply with applicable AI regulations

Little to no familiarity across most of the organization A handful of enthusiasts use AI tools; most employees do not Many employees use AI tools regularly with informal training Widespread, confident use supported by formal training programs

Leadership has shown little interest or engagement Leadership is curious but has not committed resources Leadership sponsors specific initiatives and allocates budget AI adoption is a stated strategic priority with visible executive sponsorship

No relevant in-house talent today One or two people have relevant skills as a side responsibility A small dedicated team exists but is stretched thin A well-staffed team with the skills to build, deploy, and maintain AI solutions

Past process changes have met significant resistance and stalled Change happens slowly and requires heavy management intervention The organization adapts reasonably well with proper communication A strong culture of continuous improvement makes adoption straightforward

None identified yet A few ideas discussed informally, none prioritized A prioritized shortlist exists based on value and feasibility A living, prioritized portfolio is reviewed and updated regularly

No pilots or deployments to date One pilot attempted, but not yet in production or since abandoned One or more pilots are live in production at limited scale Multiple AI solutions are live in production and delivering measured value

We have no way to measure impact today Impact is assessed anecdotally without hard metrics Defined metrics exist for major initiatives ROI is tracked systematically and informs ongoing investment decisions

Use cases are experimental with no clear link to strategy Some alignment exists but is not deliberate Use cases are chosen to support specific strategic goals Use case selection is tightly integrated into strategic planning

Legacy, on-premise systems with limited flexibility A mix of legacy and cloud systems with significant integration debt Largely cloud-based with reasonable integration between systems Modern, cloud-native, API-first infrastructure built for rapid integration

Core systems have no meaningful integration capability Integration is possible but requires significant custom engineering Modern APIs make most integrations straightforward A robust integration layer supports rapid, low-effort connections

No specific controls beyond general IT security Basic access controls exist but no dedicated monitoring Access is controlled and usage is monitored for major tools Comprehensive controls, monitoring, and DLP cover all AI tool usage

Would require major re-architecture to scale Could scale with meaningful investment and lead time Could scale with moderate effort using existing cloud capacity Elastic, cloud-native architecture scales with minimal friction