AI Implementation Readiness Assessment

A practical, founder-built assessment for enterprises planning AI implementation. 26 questions across 6 dimensions. Score your maturity in about ten minutes and walk away with a per-category playbook for the next 90 days.

Privacy: your answers stay in your browser. Nothing is uploaded, tracked, or sold. Bookmark the page to come back to a saved set of answers.

How it works

1. Answer. Pick the option that best describes your organisation today, not the aspiration. There are no right or wrong answers.
2. Score. The tool computes a maturity score per dimension and an overall tier: Exploring, Foundational, Operational, or Transformational.
3. Act. For each dimension, you get a single concrete next move calibrated to your current tier. No 50-page reports.

What it covers

Strategy & Leadership
Whether AI is treated as a tactical experiment or a strategic capability with executive ownership.
Data Infrastructure
The data foundation that determines whether AI projects can ship reliably or stall at the proof-of-concept stage.
AI Expertise & Talent
Whether you have the people who can build, ship, and operate AI systems, or whether you rely entirely on vendors.
Use Cases & Initiatives
How many AI initiatives you actually have in production, and whether they deliver measurable value.
Culture & Innovation
Whether the organisation rewards experimentation, tolerates failure, and adopts new tools quickly.
Governance, Ethics & Risk
Whether AI risks (bias, privacy, security, regulatory) are managed systematically or treated as someone else's problem.

0 of 26 answered0%

1. Strategy & Leadership

Whether AI is treated as a tactical experiment or a strategic capability with executive ownership.

Q1.1Executive sponsorship for AI initiatives.
Executive sponsorship for AI initiatives.
1No named sponsor. AI is bottom-up and exploratory.2Sponsorship from a single line-of-business leader, no enterprise mandate.3Named executive sponsor with dedicated budget and quarterly reviews.4AI is owned at the C-suite (CTO, Chief AI Officer, or equivalent) with board-level oversight.
Q1.2Documented AI strategy aligned with business goals.
Documented AI strategy aligned with business goals.
1No formal AI strategy.2Draft strategy under discussion, not yet endorsed.3Documented strategy with measurable KPIs and a 12-24 month roadmap.4AI strategy is fully integrated into the corporate strategy and refreshed every planning cycle.
Q1.3Budget allocated specifically to AI in the current fiscal year.
Budget allocated specifically to AI in the current fiscal year.
1No dedicated AI budget. Costs absorbed ad hoc by IT.2Pilot-scale budget (under 1% of IT spend) for experiments.3Multi-million budget for production AI workloads and platform investment.4AI is a top-3 line item in the technology budget with multi-year commitment.
Q1.4Board and senior-leadership AI literacy.
Board and senior-leadership AI literacy.
1Leadership has limited understanding of AI capabilities or risks.2Leadership receives occasional briefings, mostly vendor-led.3Regular leadership education programme, internal or external.4Board includes AI-fluent members and reviews AI risk and progress in every meeting.
Q1.5Tie between AI initiatives and P&L outcomes.
Tie between AI initiatives and P&L outcomes.
1AI initiatives are not tied to revenue or cost metrics.2Some projects have business cases but outcomes are not tracked.3Active projects have measurable revenue, cost, or risk-reduction targets.4Every funded AI project has an owned P&L line and is sunset if it underperforms.

2. Data Infrastructure

The data foundation that determines whether AI projects can ship reliably or stall at the proof-of-concept stage.

Q2.1Data quality and accessibility for AI use cases.
Data quality and accessibility for AI use cases.
1Data is siloed, inconsistent, and requires manual extraction.2Some domains have clean data; others require significant prep before use.3Most critical data is well-governed, documented, and accessible via API.4Enterprise-wide data catalogue with lineage, quality scores, and self-service access.
Q2.2Centralised data platform (lake, lakehouse, warehouse).
Centralised data platform (lake, lakehouse, warehouse).
1No central platform. Data lives in individual systems.2Platform exists but is partially adopted across business units.3Production data platform with most business units onboarded.4Mature platform powering both analytics and ML workloads at scale.
Q2.3Real-time data availability for AI inference.
Real-time data availability for AI inference.
1Real-time data is not available; everything is batch-only.2Limited real-time data for a few systems via custom pipelines.3Event-streaming platform (Kafka, Kinesis, Pulsar) supports several use cases.4Real-time data is the default; batch is the exception.
Q2.4Data governance and quality monitoring.
Data governance and quality monitoring.
1Informal governance; data quality issues are discovered by users.2Some policies exist but enforcement is uneven.3Formal data governance programme with named owners per domain.4Automated quality monitoring, SLA-backed datasets, regular audits.
Q2.5Unstructured data (documents, images, audio, logs) readiness.
Unstructured data (documents, images, audio, logs) readiness.
1Unstructured data is not catalogued or used for AI.2Some teams parse documents or images ad hoc.3Vector store or document-AI pipeline in production for at least one use case.4Mature pipeline indexing all major unstructured sources into a queryable store.

3. AI Expertise & Talent

Whether you have the people who can build, ship, and operate AI systems, or whether you rely entirely on vendors.

Q3.1In-house AI / ML engineering capacity.
In-house AI / ML engineering capacity.
1No in-house AI engineers. Everything is outsourced or unbuilt.2A few engineers with side AI projects but no dedicated AI team.3Dedicated AI / ML team of 5-20 people shipping production systems.4Multiple AI engineering teams aligned to business units with platform support.
Q3.2Data science / applied research depth.
Data science / applied research depth.
1No data scientists or applied researchers.2One or two generalist data scientists, mostly working on dashboards.3Embedded data scientists in key business units with applied ML deployments.4Applied research function with published work and contributions to open source or academic venues.
Q3.3Upskilling and AI literacy for non-AI staff.
Upskilling and AI literacy for non-AI staff.
1No formal training. Staff learn on their own.2Occasional lunch-and-learns or vendor training.3Structured upskilling programme with role-specific tracks.4AI literacy is part of onboarding and ongoing performance reviews.
Q3.4External partnerships and ecosystem.
External partnerships and ecosystem.
1No active partnerships with AI vendors, consultancies, or universities.2Vendor relationships exist but are mostly transactional.3Strategic relationships with two or three AI vendors and one academic partner.4Multi-vendor ecosystem with co-development arrangements and active university research collaborations.

4. Use Cases & Initiatives

How many AI initiatives you actually have in production, and whether they deliver measurable value.

Q4.1AI use cases in production today.
AI use cases in production today.
1Zero production AI use cases.2One or two pilots running in production.35-15 production AI workloads with measurable usage.4Dozens of production AI workloads spanning multiple business functions.
Q4.2Path from pilot to production.
Path from pilot to production.
1No defined process. Pilots stall or are rebuilt by the receiving team.2Informal process; some pilots reach production after long delays.3Documented pilot-to-production process with platform support.4Pilot graduation is a routine, instrumented part of the AI operating model.
Q4.3Cross-functional AI initiatives.
Cross-functional AI initiatives.
1AI is run inside a single team or function.2Occasional cross-functional projects with limited impact.3Several cross-functional AI initiatives with shared ownership and metrics.4AI is woven into nearly every cross-functional programme.
Q4.4Generative AI / LLM integration.
Generative AI / LLM integration.
1No generative AI usage beyond individual employees on consumer tools.2Sanctioned access to a GenAI tool, no in-house apps.3Internal apps built on hosted LLM APIs with guardrails and monitoring.4Self-hosted or fine-tuned models, agentic systems in production, retrieval-augmented apps at scale.

5. Culture & Innovation

Whether the organisation rewards experimentation, tolerates failure, and adopts new tools quickly.

Q5.1Employee openness to AI tools.
Employee openness to AI tools.
1Active scepticism or fear about AI replacing roles.2Mixed reception. Early adopters use AI; most do not.3Broad adoption of sanctioned AI tools across functions.4Employees actively propose AI use cases and share what works internally.
Q5.2Risk tolerance for AI experimentation.
Risk tolerance for AI experimentation.
1Low. Every experiment requires lengthy approval.2Medium. Pilots are approved but with conservative scope.3Healthy. Time-boxed experiments are encouraged.4High. The organisation expects regular failures as a sign of useful experimentation.
Q5.3Internal knowledge sharing about AI.
Internal knowledge sharing about AI.
1No forums or channels for AI knowledge sharing.2An occasional brown-bag or Slack channel.3Active community of practice with regular events and shared playbooks.4Internal conferences, fellowship programmes, and dedicated time for AI exploration.
Q5.4Change management capacity for AI deployments.
Change management capacity for AI deployments.
1Change management is not factored into AI projects.2Some projects include training, mostly as an afterthought.3Change management is a standard workstream on AI projects.4Dedicated change management function partners with every AI rollout.

6. Governance, Ethics & Risk

Whether AI risks (bias, privacy, security, regulatory) are managed systematically or treated as someone else's problem.

Q6.1Documented AI ethics or responsible-AI policy.
Documented AI ethics or responsible-AI policy.
1No policy exists.2Draft policy or aspirational principles, not yet enforced.3Documented policy with review processes for new projects.4Mature responsible-AI programme with named accountability and external audit.
Q6.2Bias, fairness, and explainability practices.
Bias, fairness, and explainability practices.
1No bias or fairness review of models.2Ad hoc review on high-profile projects.3Standard fairness and explainability checks in the model lifecycle.4Automated fairness monitoring in production with documented mitigation playbooks.
Q6.3Data privacy and regulatory compliance for AI.
Data privacy and regulatory compliance for AI.
1AI projects bypass or unintentionally violate privacy rules.2Privacy review happens late in the project, often blocking shipping.3Privacy and compliance are embedded from project intake.4Privacy-by-design is automated through tooling and platform-level controls.
Q6.4AI incident response and model monitoring.
AI incident response and model monitoring.
1No monitoring of deployed models. Incidents are discovered by users.2Basic uptime monitoring; no drift or quality monitoring.3Drift detection, performance monitoring, and a defined incident playbook.4Full AI observability stack with automated rollback, evaluation gates, and red-team programme.