A client asked me a sharp question this year, the kind that's really a test: "You're a Microsoft specialist. So how do I know you're not just going to tell me the Microsoft answer to everything?" It's a fair challenge, and my reply is the thing this whole post is about. I told her that being genuinely deep on the Microsoft stack is precisely what lets me tell her when it's the wrong choice — because I know exactly where it's brilliant, exactly where it's mediocre, and exactly where you'd be better served elsewhere. Shallow familiarity produces vendor loyalty. Real depth produces honesty. Those look similar from the outside and are opposites underneath.

Why depth and neutrality aren't in tension

The instinct behind my client's question is that specialisation and objectivity pull against each other — that the more Microsoft-deep I am, the more Microsoft-biased I must be. I understand the instinct and I think it's exactly backwards, and it's worth explaining why, because the reasoning is the whole of my professional position.

Someone with shallow knowledge of a platform is the one who oversells it. They've seen the demos, read the marketing, built a couple of things that worked — and they have no map of the edges, so they recommend it for everything, because from where they stand it looks like it does everything. It's the person who's only ever used a hammer who thinks every problem is a nail.

Deep knowledge is different. When you've actually built the hard things, hit the real limits, been burned by the specific weaknesses, you develop a precise internal map: here it's genuinely best-in-class, here it's fine, here it's a compromise, and here you should honestly use something else. That map is only available through depth. So the deeper I've gone on Microsoft, the more able I am to say "not this, not here" — because I know where the edges are from having stood on them, not from having read where they're rumoured to be.

Where I actually tell clients to look elsewhere

Vague claims of neutrality are cheap, so let me be concrete about the kinds of moment where I've told Microsoft-centric clients that Microsoft isn't the answer:

  • When the ecosystem gravity isn't there. A lot of Fabric's real value is how tightly it wires into the rest of the Microsoft world — identity, Office, Teams, the whole estate. For an organisation deep in that world, that gravity is a genuine benefit. For one that isn't — a client living mostly outside the Microsoft estate — that same benefit largely evaporates, and I'll say so, and point them at whatever actually fits their world instead of dragging them into ours for the sake of a logo.
  • When a specialist tool is simply better at the one thing that matters. For certain heavy data-engineering and data-science-first workloads, I've told clients that a specialist platform will serve them better than the generalist Microsoft option, and that the right architecture is to use that specialist tool and integrate it — not to force the Microsoft product to be something it isn't as good at being. I've argued the same about not migrating everything onto one platform just because you can.
  • When "good enough and already here" beats "better and new." Sometimes the honest advice is to not adopt the shiny Microsoft thing at all, because what they already run does the job and the migration cost buys nothing. The most vendor-neutral sentence I say regularly isn't "use a competitor" — it's "change nothing; you don't have this problem."

Why holding this line is good business, not just good ethics

Here's the part that makes this sustainable rather than merely noble: telling clients the honest answer, including "not Microsoft" and "not yet," is the best business decision I make, not a sacrifice I make despite the business.

The moment a client catches you recommending your specialism reflexively — pushing the thing you sell regardless of fit — every subsequent recommendation is discounted. They can no longer tell your genuine "yes, this fits" from your reflexive "yes, I sell this," so they have to second-guess all of it. But the advisor who has visibly, repeatedly said "don't use this here, use that instead" has banked something priceless: when that advisor says "yes, Microsoft is genuinely right for this," it lands with full weight, because it's clearly a judgement and not a sales reflex. My "not this" is what gives my "yes, this" its value. The honesty isn't in tension with the specialism — it's what makes the specialism bankable.

A specialist who can't tell you when their specialism is the wrong choice isn't a specialist — they're a salesperson with better vocabulary. Depth you can trust is depth that knows its own edges and will name them out loud, even when naming them costs you the sale.

The position, stated plainly

So I hold two things that the market keeps insisting are contradictory, and aren't. I am Microsoft-deep — genuinely, hands-on, unfashionably specialised, and I don't apologise for it, because depth is where real credibility lives. And I am vendor-neutral — not in the wishy-washy "I have no opinions" sense, but in the demanding sense that I'll tell you exactly when the thing I know best is the wrong tool for your job, and where to go instead. Those aren't a compromise between each other. They're the same commitment seen from two sides: to know a thing so well that I can be completely honest about it, edges and all. That's the whole offer. If I ever start recommending Microsoft for everything, fire me — because it'll mean I've stopped being deep enough to know better, and started being shallow enough to just sell.