Perspectives on the
intelligence era.

The pattern behind failed AI deployments is consistent — and it is not about the technology. Organisations that struggle to extract lasting value from AI share a common characteristic: they treated adoption as the destination rather than the starting point.

The installation trap

Most AI initiatives begin with a legitimate observation: a new tool exists that could help the organisation work faster, smarter or more efficiently. A decision is made to adopt it. Licences are purchased, accounts are created, training sessions are held. And then, six months later, usage drops, enthusiasm fades and the initiative is quietly deprioritised.

This is not a technology failure. It is a capability failure. The organisation installed the tool but did not build the capability to use it effectively — and those are two entirely different things.

What capability actually means

Capability is not the same as access. Giving every employee access to an AI tool does not create capability any more than giving every employee a spreadsheet creates financial analysis capability. Capability requires three things that tools alone cannot provide:

• Structured habits — individuals need to develop consistent ways of working with AI, not just experiment with it occasionally
• Redesigned workflows — the work itself needs to change, not just the tools used to do it
• Governance — someone needs to remain accountable for the outputs AI produces

The readiness question

Before any organisation invests further in AI tools, it is worth asking an honest question: at which layer of the DDITS framework does your organisation actually sit? Not where you aspire to be — where you genuinely are today.

Most organisations actively deploying AI tools are still at Layer 1 or Layer 2 — individual usage and early work redesign. That is not a failure; it is a starting point. The problem arises when organisations believe they are at Layer 4 or 5 because they have purchased enterprise AI subscriptions, when the actual capability building has barely begun.

The one thing that changes everything

The organisations that successfully build lasting AI capability share one characteristic: they treat intelligence transformation as a structured journey rather than a one-time deployment. They move deliberately through each layer, building the habits, workflows, leadership practices and governance that make each stage sustainable before advancing to the next.

The DDITS Readiness Self-Assessment is designed to help you identify exactly where your organisation sits today — and what to build next. It takes less than ten minutes and the result is a specific, actionable recommendation rather than a generic score.

Download the free self-assessment →

Most organisations overestimate their AI maturity. This is not dishonesty — it is a natural consequence of measuring adoption rather than capability. When the question is "are we using AI?" rather than "have we built the capability to use it well?", the answer is almost always more optimistic than the reality warrants.

The seven DDITS layers each have specific readiness signals — observable things that are either true of your organisation or not. They are not about aspiration or investment. They are about what is actually happening on the ground.

How to read the signals honestly

Work through each layer from 1 to 7. Your organisation's current position is the first layer where you cannot honestly tick every signal. That is your ceiling — and it is where your next investment of attention belongs.

Layer 1 — Individual Intelligence (PIOM)

• Employees across different functions are independently experimenting with AI tools — not just a handful of enthusiasts
• People are developing personal prompt habits and critically evaluating AI outputs rather than accepting them uncritically
• Individual productivity improvements are visible and being talked about — faster research, better first drafts, quicker analysis

Layer 2 — Work Design (CWR)

• Leaders have mapped which tasks in which roles are repetitive cognitive work versus judgment-intensive — and the distinction is documented, not just understood informally
• At least one team has formally restructured how its work is organised in light of what AI can now handle
• Measurable productivity improvements can be attributed to the restructuring, not just to individual tool use

Layer 3 — Collaboration (HMC)

• Teams have defined which tasks are machine-led, which are human-led and which are collaborative — and these definitions are documented and shared
• Clear handoff points exist between AI-generated outputs and human review — people know exactly where their judgment is required
• Staff feel empowered to challenge, refine and override AI outputs. They are not deferring to AI out of uncertainty or inertia

Layers 4 through 7

The diagnostic card covers all seven layers with the same specificity. The pattern is consistent: each layer has three observable signals and one readiness trigger that tells you when you are genuinely ready to advance.

The honest answer to "where are we?" is almost always one or two layers below where leadership believes the organisation to be. That gap is not a problem — it is information. And information is where transformation begins.

Download the Seven-Layer Diagnostic Card →