The Mirror Test

---

This is the sixth edition of Enable Great Conversations; a series unpacking the leadership challenges behind technology decisions, exploring how clarity and confidence can be built through open conversation and experience.

---

Microsoft has moved faster than most in making AI assistants practical for enterprise use. Copilot naturally sits at the centre of that effort, bringing generative AI directly into the tools people use every day, grounded in an organisation's own data and context.

It’s a topic that’s dominating leadership discussions across industries. Many organisations are exploring enablement, asking what Copilot success looks like, and trying to avoid the mistakes they’re hearing about from other early adopters. These are valuable questions - not about features or licensing, but about readiness. The real question many are asking is this: is our organisation really ready to get value from Copilot?

This edition looks at the specific challenge that Copilot readiness raises: what it requires, where organisations typically struggle, and what the path to genuine value looks like. Getting this right matters - not just for productivity gains, but because Copilot adoption reveals something deeper about organisational maturity and the quality of leadership that shapes it.

Switching On vs Starting Well

We've seen organisations switch on Copilot well before asking or considering whether they're ready for it to be useful. The rationale for doing so is compelling: AI assistance for every employee, productivity gains across the board, innovation at scale. The decision feels straightforward: purchase licenses, roll out access, measure adoption.

Except it doesn’t take long for questions to start surfacing (believe me, we’ve seen it!). Why aren't people using it effectively? Why are the results inconsistent? Why does it sometimes surface the wrong information, or worse, information people shouldn't have access to?

The answer is rarely related to the AI itself, it's about everything that came before.

Copilot success doesn't start with licensing. It starts with the quality of your cloud foundations, the coherence of your data, and the maturity of your governance.

Copilot doesn't create capability. It reflects it.

Foundations That Matter

Copilot is built on Microsoft Graph - the intelligence layer that connects your organisation's data across Teams, SharePoint, OneDrive, and Outlook. It uses that data to understand context, surface insights, and generate responses.

This is powerful when the underlying structure is sound. But when it's not, Copilot becomes a mirror that reflects every gap in your information architecture.

Consider what happens when data quality is poor: files scattered across dozens of SharePoint sites with inconsistent naming, content duplicated in personal OneDrive folders and team drives, and metadata that was never applied (or applied incorrectly). Copilot can't generate meaningful insight from messy foundations - it just surfaces the mess faster.

Let’s also talk about identity and access. If your controls are loose, Copilot might (correction… will!) expose documents that users shouldn't see. If they're too restrictive, it can't access the information needed to provide useful answers. Either way, the problem isn't Copilot, it's the permissions structure you've been deferring decisions about for years.

Then there's governance. Sensitivity labels that were never implemented, retention policies that don't match business needs, and data lifecycle management that exists in policy documents but not in practice. Copilot doesn't fix these gaps, it makes them more expensive to ignore.

The organisations struggling with Copilot adoption aren't failing at AI. They're discovering what their cloud migration and data projects left unfinished.

The Hidden Work

This is where the real work sits - quiet, unglamorous, but absolutely essential.

Before Copilot can be truly effective, organisations need to clean up how information is structured and stored. It means auditing SharePoint architectures that grew organically without strategy, rationalising Teams structures in Microsoft 365, reviewing permissions across hundreds or thousands of sites, and configuring Microsoft Purview to provide visibility into where sensitive data lives and how it's being used.

It means establishing governance frameworks that create trust without creating friction. Policies that teams can follow because they make sense (or are systematic in nature), not because compliance demands it.

None of this is visible to end users. None of it generates excitement in board presentations. But it is the difference between Copilot as a productivity tool and Copilot as another underutilised subscription.

I've worked with organisations that spent months on this foundational work before rolling out AI broadly. The result? Higher adoption, more consistent value, and fewer governance incidents. The work wasn't glamorous, but it was strategic, and it was incredibly valuable.

The Mirror

Beyond Copilot specifically, AI readiness is a mirror.

It reveals whether your cloud strategy was strategic, or whether it was just infrastructure modernisation framed as "transformation". It shows whether your data governance approach is built on principle or assembled from vendor recommendations. It exposes whether your information architecture reflects the way your organisation actually works, or whether it's just a collection of sites and folders that grew without intention…

The organisations that moved to cloud quickly - celebrating migration milestones without addressing these deeper questions - are now discovering the cost of that speed. Copilot (and AI more broadly) isn't failing them. It's revealing what was always there.

The leaders who get this right understand that AI adoption is downstream of cloud maturity, data coherence, and governance discipline. They know that the real work happens long before the first license is activated.

Getting It Right

The organisations that are succeeding share common characteristics.

They've treated their Microsoft 365 estate as information architecture, not just collaboration infrastructure. They've made intentional decisions about how data should be structured, accessed, and governed - and they've implemented those decisions consistently.

They've invested in Purview not as a compliance checkbox, but as a foundation for trust. They've deployed sensitivity labels that reflect actual business risk. They've built lifecycle policies that make sense to the people using them.

They've aligned their IAM strategy with the way work actually happens. Permissions reflect roles. External sharing has guardrails. Information flows to the people who need it without exposing it to those who don't.

Most importantly, they've established shared ownership. IT provides the framework and tools, but the business defines what matters and how it should be protected. leadership sets clear principles that guide decisions rather than deferring them.

The outcome isn't perfect AI. It's AI that's useful, trusted, and sustainable. Adoption happens because the foundations support it. Value accumulates because the structure enables it.

The Leadership Question

This creates a decision point for every leader evaluating readiness:

You can rush to enable Copilot and hope the foundations hold. Some do this because competitors are making noise about their deployments, some do it because boards are asking about AI progress, and some do it because the promise of productivity gains feels too compelling to delay.

Or, you can treat AI readiness as what it actually is: a test of whether your cloud and data strategy is (or was ever!) coherent.

That means slowing down to assess honestly. Looking at your SharePoint / Teams architecture and asking whether it would support intelligent search, reviewing your permissions model and questioning whether it's fit for an AI context, and examining your governance frameworks to determine whether they create confidence, or serve the ticking of a box…

The mature choice isn't to rush enablement. It’s to use AI readiness as an opportunity to strengthen foundations that should have been strengthened already.

It’s also not about perfectionism. It's about alignment. Getting the structure right enough that AI can amplify capability rather than expose gaps.

In Practice

Readiness takes longer than most organisations expect, but far less time than dealing with the consequences of getting it wrong.

It’s quiet, deliberate work: auditing data, tightening permissions, implementing governance that people can follow. It won’t generate boardroom excitement, but it is what separates AI that transforms, from AI that simply creates new problems to manage.

The organisations that understand this aren’t chasing early adoption metrics; they’re measuring the way that AI is improving decision quality, reducing friction, and creating value that compounds over time.

Reality

Copilot isn't the start of your AI journey. It's the reveal of how far you've come.

The organisations that prepared - that built coherent information architecture, implemented thoughtful governance, and created trust through transparency - are seeing AI accelerate what was already working. The ones that rushed are discovering how much work remains.

AI readiness isn't about buying the right tools, it's about building the discipline to use them well and the leaders who grasp this distinction are the ones who'll extract real value from AI. Not because they moved fastest, but because they moved with intention and understood that the unsexy work of cleaning data, tightening governance, and aligning structure is what makes AI transformative.

The question now is simple: are you ready to do that work, or are you hoping AI will do it for you?

For leaders who are, the difference often lies in finding colleagues and partners who understand that readiness isn’t a checklist - it’s a mindset built on clarity, governance, and trust.

---

Enable Great Conversations

The best decisions don't happen in isolation. They happen in conversation - with trusted peers, experienced advisors, and teams who know what it’s really like.

That's what Enable Great Conversations is about: a series exploring the real moments – the ambiguous ones, the uncomfortable ones, the ones that don’t fit neatly in a playbook - where leadership is tested, and clarity is found. Each release aims to capture a single insight, decision, or challenge that helps move organisations from noise to clarity.

There are many more of these moments worth unpacking and we’ll continue to explore them in the weeks and months ahead. We hope you’ll follow along, or join the conversation in the comments below, or follow along via the Enable Great page.