Data for AI

For twenty years, getting the data ready meant the warehouse. You modelled your entities, built pipelines into them, and analytics ran on the clean thing at the end. That work still matters. It is not what an AI assistant runs on.

An assistant answers by finding things and reading them. It reaches into documents, mailboxes, team sites, the PDF somebody attached to a ticket in 2019. Nobody wrote a schema for that material. Most of it has no owner and no metadata beyond a filename.

So the question became harder. Can a machine find the right thing, is it allowed to, and can it tell that what it found is still true? Almost every disappointing rollout I have seen fails on one of those. Very few fail on the model.

Three of those four announce themselves. A missing source leaves an obvious hole. A permissions failure produces an incident and everybody hears about it. A retrieval failure returns nothing, which is irritating but honest.

The third one does not. Stale content produces a fluent, confident, wrong answer, handed to somebody who had no reason to doubt it. No error, no alert, no flag. They act on it, and the cost turns up weeks later somewhere that looks unrelated.

And here is the part worth sitting with. Every control in the Microsoft stack governs who can see what. Sensitivity labels, access reviews, discovery controls, data loss prevention — all of them are about access. None of them ask whether a document is still correct. A superseded file with immaculate permissions passes every check you own and then grounds a wrong answer with complete confidence.

That gap is where I do most of my useful work, because it is the one nobody has a product for.

People talk about data readiness as one thing that scales. It is two different problems, and treating a small company like a scaled-down large one wastes most of the effort.

A large organisation has a governance problem. Too much is visible. Years of sites with permissive sharing defaults, permissions nobody has audited, and an assistant that will faithfully surface anything a person already had access to and never noticed.

A smaller organisation usually has the opposite problem. Not too much visible, too little. The knowledge that actually runs the business sits where an assistant structurally cannot reach it, and no amount of permission tidying will help.

The evidence for the second one is stronger than most people expect. Xero has about 1.32 million UK customers and there is no Microsoft-built Copilot connector for Xero, Sage or QuickBooks. The only finance connectors are capital-markets data providers. WhatsApp reaches around 90% of UK online adults and Copilot has no way to see it at all. Meanwhile smaller firms run more parallel platforms than large ones, not fewer.

Which one you are decides where the work starts. A large organisation starts by fencing. A smaller one starts by feeding. Same method, opposite first move, and getting it the wrong way round wastes the first month.

The failures are consistent enough to list. That is the good news, because it means they are known problems rather than mysteries.

• Permissions nobody has audited. The assistant is not overstepping. It is exercising access people already had, at a speed that makes it obvious.
• Controls that do less than people think. Labelling a SharePoint site Confidential does not encrypt or restrict the files inside it. Container labels are not inherited by items, and Microsoft says so twice across two documentation pages.
• The wrong review. Entra access reviews cover group and role membership. A SharePoint site is not even a reviewable resource type there, and direct permissions on a site are exactly where oversharing lives.
• Reports mistaken for remediation. The access governance reports, the assessments, the posture dashboards all report. None of them act without a person. Deciding who should lose access is the actual project, and no product does it for you.
• Discovery controls mistaken for access controls. Some of these hide content from search results without changing who can open it. Only one of them genuinely restricts access.
• Nothing checking whether content is true. Every control above governs access. None assesses whether a document is current, superseded or simply wrong. Correctly permissioned rubbish passes every check and grounds a confident answer.

The order matters more than the tooling. Doing these in the wrong sequence is how a programme spends a year and arrives back where it started.

• Find out what it can see. Run the access reports before changing anything, and look rather than ask. Expect the first tenant-wide permissions report to take up to five days, because Microsoft documents that it does.
• Fix access before widening capability. There is always pressure to do the reverse. This is the step that gets skipped and the one that produces the incident.
• Establish which version is the live one. An owner and a review date recorded in the content, not in somebody’s head. Superseded material moved out of retrievable space rather than left for reference.
• Make the content readable. Scanned documents actually converted, tables extracted as tables. This one is measurable: against clean ground truth, the best extraction pipelines still lose roughly seven points of retrieval accuracy, and table-heavy questions suffer far worse.
• Measure retrieval on your own questions. Forty real questions with known correct answers, run regularly. It tells you whether the right document was found at all, which is a different question from whether the answer read nicely.

I start by establishing what the assistant can see today, and I look rather than ask. That usually produces the most uncomfortable meeting of the engagement. Better to have it with me than in front of a regulator.

Then access before capability, every time. I will push back when the pressure runs the other way, and I will put the reason in terms your risk people already recognise.

After that it is unglamorous content work, prioritised by what people actually ask. There is no value in perfecting an archive nobody queries. I would rather make the two hundred documents behind the top fifty questions genuinely good.

Measurement goes in early, because without it every later argument is a matter of opinion. And your team ends up able to run the checks without me, which is the only version of this that survives my leaving.