Remote work in a weekend: what broke and what held
First observations on how companies are handling the emergency shift to remote in March 2020.
Over the past few weeks I have been watching the same scene play out in different companies. A shift to remote work that would normally take months of planning is happening in a matter of days. Sometimes over a single weekend.
This is a stress test that nobody planned for. And it is already showing where companies have real infrastructure - and where they have an illusion of readiness.
What breaks first
The first thing that surfaces is VPN. Most corporate VPN solutions were designed for the scenario of "a few employees on a business trip". When the entire office connects from outside simultaneously, the channel saturates, latency spikes, people drop off. This is not a security problem first - it is a capacity problem.
Second is the dependency on internal resources. File servers inside the perimeter that worked perfectly in the office suddenly become a bottleneck. If an employee was used to opening a file directly over the network, doing the same over VPN becomes painful.
Third is communication tools. If the company has no established tool for video calls and asynchronous work, the first days turn to chaos: someone calls by phone, someone writes in a messenger, someone waits for email.
What holds up
Companies that have moved to cloud tools over the past few years - email, documents, communications - handle the transition much more easily. When email is in the cloud and documents are in the cloud, the physical location of work does not matter.
Companies with an existing culture of asynchronous communication - tasks in a tracker, documentation in a wiki, processes in tools rather than in people's heads - also cope faster. Remote work exposes every informal process that only worked because people were sitting next to each other.
What this means for IT infrastructure
A few observations that are already coming up in conversations with directors:
The perimeter security model is cracking. The concept of "inside the perimeter - trusted, outside - not" worked when all employees were in the office. That is no longer true. A different approach to access management and authentication is needed.
Cloud tools are no longer a matter of preference. They are a matter of operational resilience. Companies that had been delaying the shift - for regulatory, cultural, or inertial reasons - are now seeing the cost of that delay.
Process documentation is not bureaucracy. It is a backup copy of the knowledge that previously lived in informal agreements between people in the same office. When the office is gone, those informal agreements are gone too.
First steps
If a company is in the middle of this transition right now, a few priorities:
- Make sure critical communication tools work and are adopted by everyone - not ten different ones, but one or two.
- Check VPN capacity and assess which systems genuinely require connecting through it.
- Identify people at risk of losing connectivity - employees without adequate home internet, without the right equipment.
- Make explicit the key daily processes: who, what, when, how to signal problems.
This is not the final architecture of remote work. It is what needs to be done right now, while the acute phase is ongoing.