Sovereign cloud: same dependency, different accent, premium price

Let's call it flag-washing.

Everyone wants to talk about where the data lives. Which country. Which region. Which domestic provider. Which sovereign offering. Which reassuringly local data centre with the right flag on the homepage, the right branding in the pitch deck, and the usual blue LED lighting that communicates "trust us, we have very atmospheric infrastructure."

All of this sounds comforting.

It also avoids the only question that actually matters:

Can any one provider reconstruct the whole thing?

Because if the answer is yes, congratulations — you don't have sovereignty. You have dependency in a local costume, being sold to you at a sovereignty markup.

The market, bless it, is very keen to keep gliding past this point.

A lot of what gets sold as "sovereign" today still runs on the same basic architecture: one provider, one control plane, one environment, one place where the full dataset can be assembled, indexed, exposed, compelled, breached, misconfigured, handed to a regulator, handed to a hacker, or quietly redefined in a terms-of-service update that nobody reads until it matters enormously.

But apparently this is fine now because the rack is nearby and there are more local nouns in the contract.

Very reassuring. Very sovereign. Extremely not the point.

The issue is not that location doesn't matter. It does.

The issue is that location is only part of the problem, while concentration is usually the part that ruins your afternoon, your quarter, and eventually your relationship with your board.

If one provider can reconstruct the full file, then one breach matters. One subpoena matters. One rogue insider matters. One misconfiguration matters. One outage matters. One commercial dispute matters. One policy change matters. One acquisition of your provider by a company you would not have chosen matters enormously.

That is not sovereignty.

That is a single point of leverage in a patriotic sweater, sold by a vendor who has done excellent work on their About Us page.

This is why I built SkyeConnex.

Not to create another "trust us, but locally" cloud story.

Not to participate in sovereignty theatre.

Not to help organizations feel marginally better about the same architectural dependency, just with a more geographically flattering invoice and a vague sense that the flag is doing protective work it was never designed to do.

SkyeConnex starts from a different premise:

No single provider should ever hold the whole file, the whole risk, or the whole leverage.

That is not a feature. That is the architecture. We distribute data across environments so that no one provider ever possesses the complete usable asset.

Which changes the conversation from "which cloud should we trust with everything?" to "why are we giving any one cloud everything in the first place?"

That second question is better. It is also significantly less convenient for a lot of marketing departments, which is probably why it keeps getting quietly skipped.

Real sovereignty is not about where data rests.

It is about who can reconstruct it. Who can compel it. Who can inspect it. Who can lose it in one clean motion. Who becomes your choke point the moment things get legally, commercially, or geopolitically complicated.

Too much of the current market treats sovereignty as a geography exercise.

Put the data in-country. Hire local staff. Add jurisdiction-friendly language. Give the offering a name that sounds somewhere between constitutional and vaguely therapeutic. Commission a press release that uses the word "resilience" fourteen times. Done.

Geography matters. I'll say it again so nobody misquotes me — geography matters.

But geography alone does not reduce concentration risk. It does not eliminate single-provider exposure. It does not change who holds technical power if one platform still has the complete dataset. And it does not solve the fundamental architectural flaw of putting the whole crown jewel in one place and then staging a series of quite thoughtful conversations about sovereignty while the blast radius sits unchanged.

You do not get sovereignty because the rack is nearby.

You get closer to sovereignty when no single provider can reconstruct what matters.

Here is the test. It is brutally simple. Use it in your next vendor meeting and watch what happens:

"Can you, as a single provider, reconstruct our full dataset?"

If the answer is yes: you are concentrated. If you are concentrated: you are exposed. If you are exposed: calling it sovereign is a very confident creative decision that I respect enormously and disagree with completely.

The shift the market needs to make is this:

From residency to reconstructability. From location to control. From branding to architecture. From "where is it hosted?" to "who can actually do something with it when something goes wrong?"

That is the lane SkyeConnex is in.

Not the "same concentration, different brochure" category that is currently quite crowded.

The much less fashionable, significantly more useful category of making sure no single provider has enough of the puzzle to become the leverage. Which, in my view, is a better definition of sovereignty than the one the market keeps selling — and buying, which is somehow the more confusing part.

Less theatre. More architecture.

Awkward, I know. We have brochures anyway.

Originally published by Ross Norrie, founder of SkyeConnex, on LinkedIn.

Published April 9, 2026 · More from the SkyeConnex blog