AI Readiness Pulse

Thirty questions, eight pillars, five maturity levels and your next 90 days. Free, in the browser, and nothing you answer leaves it.

AI Readiness Pulse

How ready is your organisation for AI at scale?

Thirty questions across eight pillars: strategy, processes, information, technology, governance, people, scaling and value. About ten minutes. Your result appears at once, and nothing you answer leaves your browser.

30questions
8readiness pillars
5maturity levels
10minutes, about

Before you start

Answer for todayChoose how the organisation works now, not the target state.
Use "Not sure"Uncertainty is useful. It is reported separately instead of forcing a false score.
An indication, not an auditA full assessment later checks the answers against evidence and with more people.

Section 1 of 8

Is AI anchored to business priorities and executive ownership?

We start with direction. Strong AI programmes know which business outcomes matter, who is accountable, how opportunities are prioritised and how investment follows value.

Strategy and Leadership · 4 questions

AI Strategic AmbitionHow clearly is AI linked to your business strategy?

Choose the statement closest to how your organisation works today.

AI is mainly experimental and not linked to a shared business agenda.Some teams have AI goals, but they are not aligned across the organisation.Enterprise AI objectives are defined and linked to business priorities.AI objectives and KPIs actively guide portfolio and investment decisions.AI strategy continuously adapts based on business results, market change and new capabilities.Not sure, I do not have enough visibility

Executive Sponsorship and AccountabilityWho is accountable for AI outcomes at executive level?

Choose the statement closest to how your organisation works today.

There is no named executive owner for enterprise AI outcomes.Sponsorship exists for individual initiatives, but accountability varies.A named executive sponsor has defined accountability for enterprise AI outcomes.Executive governance actively steers AI performance, risk and investment.AI accountability is embedded in enterprise governance and continuously reviewed as the business evolves.Not sure, I do not have enough visibility

AI Portfolio and PrioritisationHow are AI opportunities prioritised across the organisation?

Choose the statement closest to how your organisation works today.

AI ideas are mostly pursued ad hoc.Teams maintain separate lists and prioritise locally.A common portfolio uses transparent criteria to compare AI opportunities.The portfolio is actively prioritised using value, risk, feasibility and capacity.The portfolio is continuously rebalanced using realised results and strategic change.Not sure, I do not have enough visibility

Investment and Resource AllocationHow is AI investment funded and allocated?

Choose the statement closest to how your organisation works today.

Funding is mostly unplanned or taken from local discretionary budgets.AI is funded project by project with limited enterprise coordination.AI funding is planned through an enterprise budget or investment process.Funding is allocated according to portfolio priorities and measurable value.Capital is dynamically reallocated toward the highest-value AI opportunities.Not sure, I do not have enough visibility

Section 2 of 8

Can you redesign work around AI rather than simply add tools?

This section looks at process visibility, AI opportunity analysis, human-AI responsibilities and measurable baselines before automation or agentic execution is introduced.

Processes and Operating Model · 4 questions

Process TransparencyHow well are your critical end-to-end processes understood?

Choose the statement closest to how your organisation works today.

Critical processes are mostly undocumented and depend on individual knowledge.Some processes are mapped, but coverage and ownership vary by team.Priority processes are documented with owners, handoffs, systems and decision points.End-to-end processes are actively managed using common operational KPIs.Processes are continuously optimised using live performance data and governed adaptation.Not sure, I do not have enough visibility

AI Opportunity DecompositionHow systematically do you identify where AI can improve a process?

Choose the statement closest to how your organisation works today.

AI opportunities are identified mainly through intuition or isolated ideas.Selected teams analyse tasks for automation or AI support.Priority processes are broken into tasks, decisions, information needs and judgment points to identify AI opportunities.A common method uses process data to identify and prioritise AI opportunities across functions.Continuous process insight identifies and updates AI opportunities as conditions change.Not sure, I do not have enough visibility

Human-AI Operating DesignHow clearly is work divided between people and AI in future-state processes?

Choose the statement closest to how your organisation works today.

The division of work between people and AI is not explicitly designed.Human-AI responsibilities are discussed in selected pilots only.Priority workflows define which activities are human-led, AI-assisted, automated or agentic.Human control points and AI responsibilities are embedded in end-to-end process design.The division of work adapts based on performance and risk within defined governance guardrails.Not sure, I do not have enough visibility

Process Measurement and ImprovementDo you establish baseline process KPIs before introducing AI?

Choose the statement closest to how your organisation works today.

AI initiatives usually start without a measurable process baseline.Some initiatives establish baselines, but the practice is inconsistent.Priority AI initiatives use defined baseline KPIs before implementation.Baselines are linked to ongoing AI and process performance management.Continuous measurement and predictive insight drive ongoing process and AI optimisation.Not sure, I do not have enough visibility

Section 3 of 8

Can AI access trusted, contextualised and governed information?

AI depends on more than structured data. We assess ownership, quality, access, lineage and the readiness of documents, emails, contracts and other unstructured information.

Data and Information · 4 questions

Information Asset Visibility and OwnershipHow well can you identify the data and information your AI initiatives depend on?

Choose the statement closest to how your organisation works today.

Critical information is fragmented and ownership is often unclear.Selected datasets or repositories are inventoried, but coverage is incomplete.Critical structured and unstructured information assets are inventoried with accountable owners.Enterprise cataloging, ownership and access are integrated across priority AI domains.Information assets are continuously discovered, classified and governed as their use changes.Not sure, I do not have enough visibility

Quality and Fitness for PurposeHow consistently is information quality measured for AI use?

Choose the statement closest to how your organisation works today.

Quality is largely unknown or checked only when problems occur.Local quality checks exist for selected data or content sources.Quality requirements and metrics are defined for information used by priority AI systems.Quality is continuously monitored and remediation is integrated into AI operations.Automated, context-aware quality controls adapt to changing AI use and business needs.Not sure, I do not have enough visibility

Access, Integration and LineageHow do AI systems access the enterprise data and content they need?

Choose the statement closest to how your organisation works today.

Access is mostly manual, duplicated or uncontrolled.Projects use local or custom access methods with limited reuse.AI systems use governed and documented mechanisms to access required information.Standard reusable services provide governed access with lineage and monitoring.AI receives policy-aware, context-relevant information dynamically with traceable provenance.Not sure, I do not have enough visibility

Unstructured Information and KnowledgeHow ready are documents, emails, contracts and other unstructured information for AI?

Choose the statement closest to how your organisation works today.

Unstructured information is largely unmanaged or excluded from AI planning.Selected repositories are used, but quality, access and lifecycle controls vary.Unstructured information is included in AI readiness planning with defined ownership, metadata and access controls.Documents and knowledge are integrated into governed enterprise content and knowledge services for AI.Relevant content is delivered dynamically with rights, context and lifecycle controls automatically enforced.Not sure, I do not have enough visibility

Section 4 of 8

Can AI be integrated, tested and operated reliably at enterprise scale?

We look at common architecture, enterprise integration, identity and permissions, and the engineering controls required to move beyond isolated AI tools.

Technology, Architecture and Engineering · 4 questions

AI Reference ArchitectureDo you have a common enterprise architecture for AI?

Choose the statement closest to how your organisation works today.

AI solutions are selected and built independently with no common architecture.Some shared platforms or patterns exist, but teams still make local choices.An enterprise AI reference architecture defines approved components, interfaces and control points.Reusable platform services and architecture standards support AI at enterprise scale.The architecture supports multiple models and agents with dynamic optimisation and controlled interoperability.Not sure, I do not have enough visibility

Integration and Tool AccessHow are AI solutions connected to enterprise systems and tools?

Choose the statement closest to how your organisation works today.

Connections are manual, point-to-point or created separately for each use case.Some common connectors exist, but integration remains inconsistent.AI uses governed APIs, services or connectors to access enterprise capabilities.Reusable integration services make enterprise capabilities securely available to AI at scale.AI and agents can orchestrate approved enterprise services dynamically within policy guardrails.Not sure, I do not have enough visibility

Identity and Runtime SecurityHow are identities and permissions managed for AI systems and agents?

Choose the statement closest to how your organisation works today.

AI tools may use shared credentials or unclear permissions.Identity and access controls exist for selected systems but are inconsistent.AI systems and agents use explicit identities with defined permissions and least-privilege access.Identity, secrets and permissions are centrally governed and monitored across AI workloads.Permissions adapt to context and risk while remaining continuously monitored and auditable.Not sure, I do not have enough visibility

Testing, Observability and PerformanceHow is AI quality tested before production use?

Choose the statement closest to how your organisation works today.

Testing is informal or mainly technical.Projects define their own tests, with inconsistent business acceptance criteria.AI is evaluated against defined business, quality and risk criteria before production.Standard evaluation, observability and release controls are integrated into the delivery lifecycle.Evaluation is continuous and adapts as models, data, risk and business conditions change.Not sure, I do not have enough visibility

Section 5 of 8

Can you scale AI without losing accountability, control or trust?

This section tests whether AI use is visible, risk-classified, approved and monitored through controls proportionate to business and regulatory risk.

Governance, Risk, Security and Compliance · 4 questions

AI Policy and AccountabilityHow clearly are acceptable and prohibited uses of AI defined?

Choose the statement closest to how your organisation works today.

There is no enterprise policy or consistent guidance.Guidance exists in some areas, but interpretation varies.Enterprise policy clearly defines acceptable, restricted and prohibited AI use.Policy is risk-based, embedded in delivery and regularly reviewed using operational evidence.Policy and controls adapt continuously as technology, regulation and business use evolve.Not sure, I do not have enough visibility

AI Inventory and Risk ClassificationDo you know which AI systems, models and agents are in use and how risky they are?

Choose the statement closest to how your organisation works today.

There is no reliable inventory of AI use.Some teams maintain lists, but coverage and risk classification are incomplete.Material AI systems, models, agents and high-risk uses are inventoried and risk-classified.The inventory is integrated with approval, monitoring, ownership and lifecycle controls.Discovery and risk classification are continuously updated as AI use changes.Not sure, I do not have enough visibility

Approval, Testing and Human OversightHow are AI solutions approved before production?

Choose the statement closest to how your organisation works today.

There is no consistent production approval process.Approval depends on the team or project.Production AI requires documented approval proportionate to use-case risk, including defined human oversight.Approval evidence is integrated into delivery workflows and control gates.Risk-based controls are increasingly automated while preserving accountable human decisions where required.Not sure, I do not have enough visibility

Monitoring, Incident and Change ManagementHow are AI solutions monitored after go-live?

Choose the statement closest to how your organisation works today.

Ongoing AI quality and risk are rarely monitored systematically.Some production solutions are monitored, but metrics and escalation vary.Production AI has defined monitoring for material quality, performance and risk indicators.Monitoring, incidents and material changes are integrated into operational governance and service management.Continuous risk and performance signals trigger adaptive controls, escalation and improvement.Not sure, I do not have enough visibility

Section 6 of 8

Can people adopt AI effectively, and are roles evolving with it?

AI readiness depends on literacy, role-specific skills, structured change and deliberate redesign of responsibilities between people and AI.

Organisation, Skills and Change · 4 questions

AI LiteracyHow widely do employees understand AI capabilities, limitations and risks?

Choose the statement closest to how your organisation works today.

Understanding depends mainly on individual interest and experience.General awareness training exists for selected groups.A defined AI literacy baseline is provided to relevant employees and leaders.AI literacy is role-based, reinforced through work practices and measured for adoption outcomes.Learning continuously adapts as roles, technology and risk evolve.Not sure, I do not have enough visibility

Role-Specific AI SkillsAre AI skills defined for the roles that need them?

Choose the statement closest to how your organisation works today.

AI skills are not formally defined by role.Selected technical or business teams have local skill expectations.Role-specific AI competencies are defined for priority technical, business and leadership roles.Skills are linked to workforce planning, learning paths and transformation priorities.Skills intelligence continuously updates role profiles and development as work changes.Not sure, I do not have enough visibility

Adoption and Change ManagementHow is user adoption and change managed in AI initiatives?

Choose the statement closest to how your organisation works today.

Teams largely assume users will adopt the solution.Some projects include communications or training, but change practices vary.Priority AI initiatives have explicit adoption, communication and change plans with accountable owners.Adoption is measured and managed as part of transformation and benefits realisation.Change mechanisms continuously adapt using usage, workforce and performance feedback.Not sure, I do not have enough visibility

Human-AI Role RedesignHow systematically are roles and responsibilities redesigned when AI is introduced?

Choose the statement closest to how your organisation works today.

Role impacts are rarely assessed explicitly.Role changes are discussed for selected initiatives after solutions are designed.Priority AI initiatives explicitly assess changes to roles, responsibilities and decision rights.Human-AI role design is integrated into workforce planning and operating-model change.Roles continuously evolve as human-AI collaboration and business needs change.Not sure, I do not have enough visibility

Section 7 of 8

Can successful AI ideas move from experiment to production and scale?

We assess how AI ideas enter the pipeline, how validated experiments become controlled production services and how successful capabilities are reused across the enterprise.

AI Delivery, Productisation and Scaling · 3 questions

Use-Case Intake and QualificationHow do AI ideas enter the delivery pipeline?

Choose the statement closest to how your organisation works today.

Ideas proceed informally based on local enthusiasm or available budget.Some teams use repeatable intake methods, but there is no common enterprise process.AI ideas enter a common qualification process covering value, feasibility, risk and ownership.A managed portfolio uses common intake and reusable capabilities to accelerate high-priority use cases.The portfolio continuously learns from delivery and realised value to refine qualification decisions.Not sure, I do not have enough visibility

ProductionisationHow do successful AI experiments move into controlled production?

Choose the statement closest to how your organisation works today.

There is no consistent path from experiment to production.Some projects reach production, but the path depends on the team.A defined lifecycle covers production approval, ownership, operations, monitoring and support.Reusable platform and delivery patterns make productionisation predictable and repeatable.Production AI is continuously optimised as a managed product/service portfolio.Not sure, I do not have enough visibility

Enterprise ScalingHow easily can successful AI solutions be reused across teams, functions or geographies?

Choose the statement closest to how your organisation works today.

Successful solutions usually remain local to the original team.Some reuse occurs, but it requires significant rework.Standard methods and reusable components support replication across selected areas.Shared platforms, patterns and operating practices enable efficient enterprise scaling.AI capabilities are managed as reusable enterprise products with systematic scaling and retirement.Not sure, I do not have enough visibility

Section 8 of 8

Can you prove whether AI investment is creating real value?

The final section looks at business baselines, complete AI business cases and whether realised benefits are measured after go-live.

Value Realisation and ROI · 3 questions

Value Hypothesis and BaselineDoes each priority AI use case start with a clear business problem and value hypothesis?

Choose the statement closest to how your organisation works today.

AI initiatives often start from technology interest without a quantified business problem.Some initiatives define expected benefits, but baselines and ownership vary.Priority use cases define the business problem, baseline and expected value before implementation.Value hypotheses are compared across the portfolio and validated as solutions scale.Value assumptions are continuously updated using realised performance and changing business priorities.Not sure, I do not have enough visibility

Financial Business CaseHow complete are your AI business cases?

Choose the statement closest to how your organisation works today.

Business cases are absent or focus mainly on expected benefits.Some costs and benefits are estimated, but assumptions are inconsistent.Business cases include implementation, recurring operating costs, benefit assumptions and ownership.ROI, payback and value scenarios are used consistently to steer portfolio decisions.Financial assumptions and portfolio economics are continuously updated using actual operating data.Not sure, I do not have enough visibility

Benefits RealisationHow are realised AI benefits measured after go-live?

Choose the statement closest to how your organisation works today.

Success is mostly judged by delivery completion or anecdotal feedback.Some initiatives track outcomes, but post-go-live measurement is inconsistent.Realised benefits are measured against agreed baselines with accountable benefit owners.Portfolio reviews use realised value to scale, redesign or stop AI initiatives.Continuous value measurement dynamically steers AI investment and operating decisions.Not sure, I do not have enough visibility

How the result is calculated

Each answer scores from one to five, and “Not sure” scores nothing: it is reported as coverage rather than turned into a guess. A pillar is the average of its answers, and the overall score is the average of the eight pillars. No weighting and no hidden model.

The score maps to five maturity levels: Ad-Hoc below 1.8, Repeatable from 1.8, Defined from 2.6, Managed from 3.4 and Intelligent from 4.2. Two guardrails keep a good average from hiding a weak foundation. Managed needs information, governance, delivery and value at Defined or better, and Intelligent needs every pillar at Defined and processes, governance and value at Managed.

The five process dimensions (strategic, transactional, interactional, transformational and compliance) are read from the questions tagged to each. The 90-day priorities follow the weakest of the pillars that block scaling, which is where effort pays first.

The result is an indication from your own answers, not an audit. The full assessment that follows checks it against evidence and with more people from your organisation.

A score is a starting point. The conversation that follows is where it turns into a plan, and it costs nothing.

Book a conversation