Rollout & AdoptionOperational Design

Middle Management Is the Bottleneck (and the Solution) of AI Adoption

Executives sponsor the rollout. Individual contributors quietly use the tools anyway. The layer that decides whether anything actually changes sits in between, and most programmes ignore it.

AUG 2026·5–6 min read

AI Governance

August 2 Came Anyway: What the AI Act Delay Did and Didn't Buy You

The EU pushed its high-risk deadlines to 2027 and 2028, and plenty of boards read that as 'stand down'. The transparency obligations that took effect this week say otherwise.

AUG 2026·4–6 min read

Data ArchitectureAI Strategy

What AI Agents Need to Read Your Data: MCP and the New Integration Layer

The Model Context Protocol has quietly become the standard way agents connect to enterprise systems. It collapses the integration problem from M×N to M+N, and moves the hard question from plumbing to permissions.

JUL 2026·5–7 min read

Program DeliveryAI Strategy

Run Your AI Like a Portfolio: Budget in Quarters, Not Years

An eighteen-month AI roadmap is fiction by slide four. The alternative is not less discipline. It is a different discipline: small positions, quarterly rebalancing, and a hit rate instead of a Gantt chart.

JUL 2026·4–5 min read

Program DeliveryAI Governance

What Does 'Done' Mean for AI Systems?

You can't sign off a probabilistic system with a test checklist. Evals are the new acceptance tests: a golden set the business owns, error budgets, and a bar that gets set before anyone sees the demo.

JUL 2026·5–6 min read

Program DeliveryAI Strategy

Planning for a Number You Don't Have Yet

Stage-gate governance assumes you can specify the outcome before you fund the work. AI projects can't promise an accuracy figure before anyone has touched the data. Here is how to plan around a number you still have to discover.

JUL 2026·5–6 min read

AI StrategyRollout & Adoption

Digital × Human: Digital Transformation Between Technology and People

Most AI programmes do not fail on the technology. They fail in the space between a working model and a changed way of working. That gap is the whole reason DxH exists, and this is how we think about closing it.

JUN 2026·6–8 min read

AI StrategyOperational Design

Claude Fable 5: When to Use It, and When Not To

Anthropic's new top-tier model is twice the price of Opus. For a small set of workloads it is worth every cent. For most of what enterprises run, it is an expensive way to get the same answer.

JUN 2026·6–8 min read

AI StrategyOperational Design

The Claude Skills Worth Installing in an Enterprise

Skills turn a capable generalist into a model that carries your team's standards. Here are the ones that earn their place at work, and what each one actually unlocks.

JUN 2026·6 min read

AI StrategyOperational Design

Build, Buy, or Wrap? A Decision Framework for Enterprise AI

Every AI capability decision lands in the same meeting: the CTO wants to build, procurement found a vendor, and someone already wrapped a chatbot last weekend. Four questions settle the argument.

JUN 2026·8–10 min read

Rollout & AdoptionAI Strategy

The AI Literacy Gap Isn't About Training. It's About Trust.

Completion rates are high and usage is flat. The missing ingredient in most AI literacy programmes isn't knowledge. It's the safety to use what people already know.

JUN 2026·6–8 min read

Operational DesignRollout & Adoption

The Pilot Trap: Why 80% of AI Proofs-of-Concept Never Reach Production

AI pilots succeed because they are designed to succeed. Production inverts every condition that made the demo work. Here is where pilots actually die, and how to design one that can graduate.

JUN 2026·7–9 min read

AI StrategyAI Governance

The Real Cost of AI: Why the Token Bill Is the Easy Part

LLM prices keep falling, yet enterprise AI bills keep climbing. Here is where the money actually goes, why budgets disappear so fast, and what governance and change management have to do with it.

MAY 2026·5 min read

AI GovernanceAI Strategy

Shadow AI Is Already Inside. The EU Just Gave You Time to Deal With It.

Shadow AI is already inside your organisation. The EU just handed you a reprieve to do something about it.

MAY 2026·6–7 min read

AI StrategyOperational Design

Most AI Agents Should Have Been Automations (And Why That's Good News)

The market is full of AI agents that should have been simple automations. Here's how to tell the difference, when an agent genuinely earns its place, and why the answer almost always starts with the people doing the work.

MAY 2026·6–8 min read

AI StrategyOperational Design

LBO Modelling Automation in PE: A Practical Guide in 2026

From Claude embedded in Excel to purpose-built platforms like Mosaic, here's how deal teams are actually using AI to accelerate LBO modelling—and why the human in the loop is the whole point.

MAY 2026·9–11 min read

AI GovernanceAI Strategy

Why AI Governance Starts With Accountability, Not Policy

Most organisations build AI governance frameworks backwards — policy first, accountability second. When things go wrong, no one owns the outcome. Here is how to reverse that.

APR 2026·6–8 min read

Rollout & AdoptionOperational Design

Designing AI Workflows That People Actually Use

Adoption failure is rarely a technology problem. It is a workflow design problem. The teams that succeed treat AI integration as a process redesign, not a tool deployment.

MAR 2026·7–9 min read

Data ArchitectureAI Strategy

Data Architecture for AI Readiness: What Most Roadmaps Get Wrong

Organisations declare themselves AI-ready on the basis of a data lake and a handful of dashboards. Real AI readiness is an architectural property, not a data volume milestone.

MAR 2026·8–10 min read