Almost every growing business has, at some point, relied on one person who single-handedly keeps the technology running — knows every password, remembers every quirk of the network, and can fix almost anything without documentation because it's all in their head. That person is a genuine asset. They are also, structurally, a risk the business has usually never quantified.
The first failure mode is the obvious one: what happens when that person is unavailable. Not just departure — a two-week holiday, an illness, a family emergency. If the honest answer to "what happens if they're unreachable for two weeks" is "we'd struggle badly," that's a risk sitting in plain sight, not a hypothetical.
The second, quieter failure mode is scope creep without capability growth. As the business grows, this person is often asked to take on more — more systems, more complexity, more responsibility — without a corresponding increase in specialist depth in any one area, because there's no time to develop it while also keeping everything running day to day. The result is broad, shallow coverage across an increasingly complex environment, which is a worse risk profile than the same person had three years earlier, even though nothing dramatic changed.
The third is burnout, which tends to show up as exactly the kind of mistake that looks like incompetence from the outside — a missed patch, a misconfigured setting — but is actually fatigue from being the only person responsible for both strategic decisions and routine firefighting, indefinitely.
None of this is a case against generalists; it's a case for recognising the pattern before it becomes an incident. The fix isn't necessarily replacing that person — it's ensuring their knowledge is documented, their coverage has a backup, and the specialist work outside their depth has somewhere else to go.
All technical perspectives