Methodology

How We Measure AI Readiness

This assessment is not a generic quiz. It is built on a systematic literature review of peer-reviewed AI maturity models, validated against leading industry frameworks, and powered by an adaptive AI analysis pipeline.

Foundation

Framework basis

The assessment framework synthesizes findings from a systematic literature review of 13+ published AI maturity and readiness models. These were cross-referenced against four major industry frameworks to ensure comprehensive coverage and practical relevance.

MITRE AI Readiness

6 pillars and 20 dimensions covering organizational, technical, and ethical readiness for AI adoption.

MIT CISR AI Maturity

4-stage progression model (Siloed, Bridging, Modular, AI-Fueled) focusing on data architecture and organizational design.

Microsoft AI Maturity Model

Strategy, culture, organizational readiness, and capability dimensions for enterprise AI transformation.

Cisco AI Readiness Index

Infrastructure, data, governance, talent, and culture pillars benchmarked across 8,000+ organizations globally.

By triangulating academic models with practitioner frameworks, we derived a unified model with 7 dimensions built from 31 sub-criteria that avoids gaps common in single-source assessments. The resulting framework captures both technical and organizational readiness -- because AI failures are rarely just technical.

The sub-criteria define what each dimension covers. They are not a 31-question checklist: you answer eight scenario questions, one per dimension plus one on unsanctioned AI use, and the assessment then generates follow-up questions targeted at whichever dimensions look weakest. That is why two people at the same company can be asked different things -- and why the assessment takes ten minutes rather than an hour.

Framework

7 dimensions, 31 sub-criteria

Each dimension captures a distinct facet of AI readiness. The sub-criteria beneath it define what it covers and drive targeted recommendations.

Strategy & Vision

How clearly AI is embedded in your organization's strategic direction, leadership commitment, and investment planning.

Strategy Formalization -- Degree to which AI strategy is documented, communicated, and aligned with business goals.

Leadership Buy-in -- Executive sponsorship and active championing of AI initiatives.

Use Case Pipeline -- Systematic identification and prioritization of AI opportunities.

Investment Allocation -- Budget dedicated to AI initiatives relative to strategic importance.

Data Readiness

The quality, accessibility, and governance of your data assets for AI applications.

Data Quality -- Accuracy, completeness, and consistency of organizational data.

Data Accessibility -- Ability to access and integrate data across systems and departments.

Data Governance -- Policies and processes for data management, privacy, and security.

Pipeline Maturity -- Automated data pipelines, ETL processes, and real-time data capabilities.

Technology & Infrastructure

How far AI has actually reached into your systems -- from ad hoc individual use to integrated, monitored production workflows.

Model & Tool Access -- Sanctioned access to AI models and tools, whether via APIs, cloud providers, or vendor products.

System Integration -- Ability to connect AI to the systems of record where work actually happens, rather than running it alongside them.

Production Tooling -- What it takes to run AI in production: change control for prompts and configuration, output monitoring, and rollback.

Scalability -- Capacity to go from one working AI workflow to many without rebuilding each time.

People & Skills

AI literacy, talent availability, and learning culture across your organization.

AI Literacy -- General understanding of AI capabilities and limitations across the org.

Practice Depth -- Whether anyone is actually building with these tools day to day, who the internal go-to person is, and whether there is budget for tools rather than headcount.

Non-Technical Empowerment -- Business users equipped with no-code/low-code AI tools and training.

Learning Culture -- Continuous learning programs and upskilling opportunities for AI.

Governance & Ethics

AI policies, responsible AI practices, and regulatory compliance posture.

AI Policies -- Formal policies governing AI development, deployment, and usage.

Responsible AI Practices -- Controls for the failure modes of bought AI: unchecked output reaching customers, prompt injection, company data leaving to third-party models, unreviewed AI-generated code. Bias and fairness monitoring where AI touches hiring, lending, or clinical decisions.

Regulatory Compliance -- Awareness and preparedness for AI-specific regulations (EU AI Act, etc.).

Risk Management -- Processes for identifying and mitigating AI-related risks.

Agent Supervision -- Ownership of what autonomous AI is allowed to touch -- systems, spend, customer communication -- with permissions, monitoring, and a way to stop it when it is wrong.

Sanctioned Use -- Visibility of which AI tools staff actually use, whether an approved list exists, and whether anything prevents company data reaching consumer tools.

Organization & Culture

How well your organizational culture supports experimentation, innovation, and AI adoption.

Innovation Tolerance -- Willingness to experiment, accept failure, and iterate on new approaches.

Leadership Engagement -- Leaders actively participating in and supporting AI transformation.

Change Readiness -- Organizational capacity to absorb and adapt to AI-driven changes.

Cross-Functional Collaboration -- Ability of technical and business teams to work together on AI projects.

Process & Operations

How deeply AI tools are integrated into your workflows, and how you measure their impact.

Automation Baseline -- Existing process automation, which predicts how hard AI will be to integrate. Context for AI adoption rather than a measure of it -- deterministic RPA is not AI.

AI Workflow Integration -- AI tools embedded in day-to-day business processes vs. standalone experiments.

Impact Measurement -- KPIs and metrics tracking the business value of AI initiatives.

Output Verification -- How the organization knows AI output is correct: review before use, an eval set or regression suite, detection when quality drifts, and a feedback loop from the people relying on it.

Continuous Improvement -- Feedback loops from AI deployments driving iterative improvements.

Scoring

5 maturity levels

Overall and per-dimension scores map to a 5-level maturity model. Each level represents a qualitatively distinct stage of AI capability.

Exploring(0--20%)

AI is on the radar but there is no structured approach. Awareness exists but action is ad hoc.

Planning(21--40%)

Initial strategy forming. Some pilots underway but not yet systematic or scaled.

Implementing(41--60%)

Active AI projects with dedicated resources. Moving from pilots to production in select areas.

Scaling(61--80%)

AI integrated into multiple business functions. Systematic approach to scaling and measuring impact.

Leading(81--100%)

AI is a core competitive advantage. Organization-wide adoption with continuous innovation.

Assessment Design

Scenario-based assessment

Most AI readiness tools ask "how would you describe your X?" -- pure self-assessment that is easy to inflate. Ours uses behavioral interview methodology: every question describes a concrete scenario and asks what actually happened or would happen.

Observable reality, not self-assessment

Instead of asking "How mature is your data infrastructure?", we ask "If you needed to pull customer data for an AI project tomorrow, what would happen?" Each answer option describes a specific observable state -- what actually occurs in your organization, not what you aspire to.

Friction signals embedded in options

Lower-maturity answer options (levels 1-2) embed organizational friction signals: political barriers, failed attempts, active resistance. This surfaces real blockers that generic maturity labels would miss.

Conservative evaluation calibration

The AI analysis pipeline applies skepticism to self-reports. C-suite respondents tend to overreport strategic alignment; individual contributors may underreport org-level initiatives. The system accounts for these biases and scores conservatively when evidence is ambiguous.

Scoring Model

Organizational friction model

Technical readiness is gated by organizational readiness. A company with great infrastructure but toxic culture gets a middling average in traditional assessments instead of a danger signal. Our model makes this explicit.

Capability score, then organizational multiplier

The seven dimensions do two different jobs, so they are scored in two steps. Five of them -- strategy, data, technology, governance and process -- measure capability: whether you have the means to act. Their average is your capability score. Culture and people measure something else entirely: whether the organization can absorb the change. They are deliberately kept out of that average and applied afterwards as a multiplier, so neither is counted twice.

Organizational multiplier (0.40 -- 1.10)

The average of your culture and people scores maps onto a multiplier centred on 1.00, where 50/100 is neutral. It is deliberately asymmetric: friction can cut capability by up to 60%, while strong organizational health adds at most 10%. A dysfunctional organization can waste most of the capability it has; a healthy one cannot manufacture capability it lacks. Individual dimension scores stay untouched -- they reflect exactly what you answered.

Why this matters

An organization scoring 80/100 across the capability dimensions but 20/100 on culture and people cannot execute AI transformation. A flat seven-dimension average reports that as a comfortable 63 and buries the problem inside it. Here it reports a capability score of 80 dragged down to 51, and names which half is the problem. The roadmap is gated on the same logic: if culture is critically low, month 1 focuses on organizational change, not technical implementation.

Analysis

Contradiction detection

Dimensional tensions are surfaced as first-class output, not buried in prose. The system detects specific contradiction patterns between dimensions and assigns risk levels based on the gap magnitude.

Strategy-Culture Gap

Plans exist but the org can't execute. Strategy scores high while culture lags behind.

Infrastructure-Data Gap

Great tools but no usable data. Technology is advanced but data readiness is low.

Compliance Theater

Policies exist but nothing is automated. Governance scores high while process automation is low.

Skills Gap Trap

Ambitious plans but no one to execute. Strategy is high while people readiness is critically low.

Risk levels are assigned based on point gaps: high (>25 points), medium (15-25 points), low (<15 points). Only patterns that actually exist in your results are reported.

Intelligence

Adaptive assessment, not a static quiz

Most readiness assessments use the same fixed questions for every respondent. Ours adapts in real-time.

Phase 1: Scenario questions

Seven scenario-based questions (one per dimension) establish baseline scores across all 7 dimensions. These are deterministic and ensure consistent benchmarking. Every question includes an optional context field where you can add freeform notes -- the AI uses these to generate sharper analysis.

Phase 2: AI-generated follow-ups with friction probes

Claude analyzes your scenario responses in real-time and generates targeted follow-up questions. If you scored low on data governance but high on cloud maturity, the follow-ups probe the specific gap. Each adaptive question includes at least one friction probe: past failed initiatives, active resistance from teams, political dynamics blocking cross-functional work. For C-suite respondents, questions probe whether executive perception matches ground-level reality.

Review and edit

Before submission, a full review screen lets you see all answers organized by dimension. Edit any response, add context notes, or go back to earlier questions. Nothing is locked until you explicitly submit.

Why this matters

Static assessments miss nuance. A 500-person fintech and a 5,000-person manufacturer both scoring "Level 2" on data readiness have fundamentally different gaps. Adaptive questioning surfaces those differences in under 5 minutes.

Pipeline

Multi-layer analysis pipeline

After you complete the assessment, your responses flow through a three-stage analysis pipeline. A fourth stage -- interactive chat -- lets you interrogate your results with full context.

Diagnose

Scoring + friction + contradictions

Scenario answers are scored deterministically using pre-calibrated weights. Adaptive answers are scored by Claude with strict 1-5 rubrics. An organizational friction multiplier (derived from culture and people scores) mechanically adjusts the overall score. Claude detects dimensional contradictions (e.g., high strategy but low culture = plans exist, org can't execute) and generates an organizational reality assessment with specific resistance patterns and adoption barriers.

Discover

Deep diagnostic + use cases

Using your scores and raw responses, Claude generates AI use cases tailored to your industry, maturity level, and identified gaps. Each use case is ranked by impact vs effort and mapped to the dimensions where you have the most room for improvement. These are not generic suggestions -- they are grounded in what you actually told us.

Direct

Culture-gated roadmap

The final stage synthesizes your diagnostic, use cases, and organizational context into a 90-day action plan. Planning is culture-gated: if culture is critically low, month 1 focuses on organizational change, not technical implementation. Each milestone includes adoption barriers specific to your organizational friction profile. Quick wins, milestones, and scaling initiatives are calibrated to your maturity level.

Discuss

Interactive chat

After results load, an AI chat advisor is available with full context of your scores, diagnostics, use cases, and roadmap. Ask follow-up questions like "which 2 initiatives fit a $50K budget?" or "how do I pitch this to a skeptical CEO?" The chat improves as more analysis layers complete, and supports multi-turn conversations grounded in your specific assessment data.

Sources

Academic references

The following references informed the design of our assessment framework. This is a representative subset of the 13+ models reviewed.

Alsheibani, S., Cheung, Y., & Messom, C. (2018). "Artificial Intelligence Adoption: AI-readiness at Firm-Level." PACIS 2018 Proceedings. -- Foundational firm-level readiness model covering strategy, technology, and organizational factors.
Jockl, S., Fremdt, N., & Gimpel, H. (2024). "AI Readiness: A Synthesis of Existing Frameworks and the Development of a Comprehensive Instrument." Proceedings of the 57th Hawaii International Conference on System Sciences (HICSS). -- Systematic synthesis of AI readiness dimensions from multiple models into a unified measurement instrument.
Holmström, J. (2022). "From AI to digital transformation: the four stages of AI maturity and its impact on organizational competitiveness." Competitiveness Review, 32(4). -- Stage-based maturity progression and its relationship to competitive advantage.
MITRE Corporation (2023). "AI Readiness Framework." -- 6-pillar, 20-dimension framework used by US Department of Defense organizations for AI adoption assessment.
Ross, J. W., Beath, C. M., & Mocker, M. (2019). "Designed for Digital: How to Architect Your Business for Sustained Success." MIT Press. -- MIT CISR research on data architecture and organizational design as prerequisites for AI-fueled operations.
Cisco (2024). "AI Readiness Index." -- Global benchmarking study across 8,000+ organizations covering infrastructure, data, governance, talent, and culture readiness pillars.
Pumplun, L., Tauchert, C., & Heidt, M. (2019). "A New Organizational Chassis for Artificial Intelligence." ECIS 2019 Proceedings. -- Organizational design patterns for AI integration, informing the culture and process dimensions.
Lichtenthaler, U. (2020). "Five Maturity Levels of Managing AI: From Isolated Ignorance to Integrated Intelligence." Journal of Innovation Management, 8(1). -- Five-level progression model that influenced our maturity-level definitions.

See where you stand

The methodology is rigorous. The assessment is free. Takes under 5 minutes, no signup required.

Start your assessment