The conversation I keep having

Every few weeks I end up in a conversation that goes roughly like this. Someone mentions they keep project documents in their personal cloud account because it is easier to share, works on their phone, and does not require a ticket. I understand all three reasons. I’ve done the same thing myself at some point, usually late on a Friday when the official tool was being unhelpful.

The trouble is not the intent. It is that the file has now left the boundary where anyone can see it, back it up, or take it back when the person moves on. And these conversations go badly when they sound like an accusation. The moment it feels like a scolding, people stop mentioning what they do, which is worse than the original problem.

What I’ve noticed in practice

The files that end up in personal accounts are rarely random. They tend to be the ones people are actively working on: draft contracts, spreadsheets with customer names, screenshots of internal systems, exports pulled for a quick analysis. In other words the working set, not the archive.

People also have reasonable-sounding reasons. The corporate sync client is slow. The approved sharing tool has a file size limit. A client sent something to a personal address and forwarding felt easier. I’ve heard all of them, and most are legitimate complaints about our own tooling rather than carelessness.

There is a version of this I see with contractors and short-term collaborators too. They never get a managed account, so a personal drive becomes the only practical way to exchange anything. That is a provisioning gap on our side, and it produces exactly the outcome we say we don’t want.

What tends to surprise folks is the departure scenario. When somebody leaves, their personal account leaves with them. Nobody has a list of what was there, and asking for access afterwards is awkward at the exact moment it is also urgent.

Why it keeps happening

Because the personal option is genuinely more convenient in small, daily ways. It is already logged in, already on the phone, already configured for sharing with anyone. A managed workspace usually asks for more steps, more approvals and more patience.

It also happens because most people have never seen the downstream mess. They have not had to answer an access request for a file that lived in somebody else’s account. There is no feedback loop, so the behaviour continues quietly.

And, honestly, our own processes often push people there. If getting a shared folder takes three days, the workaround will keep winning, however many reminders we send. Convenience is a strong force, and it does not respond well to policy language. I’ve found that asking people to describe the last time the official tool slowed them down gives me a short, concrete list of things worth fixing, which is more useful than another awareness campaign.

What some teams try

I’ve seen a few approaches that seem to land better than policy warnings. One is simply making the sanctioned tool good enough: fast sync, phone access, external sharing that works without a support ticket. Another is giving people a short, honest explanation of what happens to files when an account disappears, with no threat attached.

I’ve also seen unified platforms such as Ping64 used to cut down the overhead of juggling many separate security tools.

The common thread is that the fix has to make the right path easier, not the wrong path scarier. Nagging has a short shelf life. Once people believe the approved option will slow them down, no amount of messaging recovers that trust.

Small practical tips

Fix the file size limit first. It sounds trivial, and it is the reason I’ve heard most often.

Make external sharing in the approved tool a one-click thing for the common case, with logging quietly running behind it.

When somebody leaves, include a plain question in the offboarding checklist: were you keeping anything work-related in a personal account? Ask it kindly, before the last day.

Offer a short amnesty window where people can move files into managed storage with no questions asked. You learn a lot from what shows up.

Talk about it in terms of continuity rather than compliance. We could not find the file lands better than you broke a rule.

Ask team leads to model the behaviour themselves. People copy what the person next to them does, not what the handbook says.

Wrap-up

I’ve come to think of this as a tooling problem wearing a behaviour problem’s clothes. Most people are not being careless, they are routing around friction. Reduce the friction, explain the continuity risk once, and the volumes tend to drop on their own. Total compliance is not really the goal; knowing where our data lives is. If I can get one team to move its working files voluntarily, that is a decent year.

References