OCI DOMAIN THREE
Network operations
This domain is rarely about whether the network works. It usually works. The question is how much manual effort is being spent keeping it that way, how much of that effort is invisible to the people funding it, and what happens when the person doing it is on leave.
Does any of this sound familiar?
Your engineers know which systems misbehave and have quietly built routines around them. None of it appears in a ticket.
Monitoring is in place. The alerts it produces are largely ignored, because most of them do not mean anything.
You know there is a single point of failure. It has been known for long enough that it no longer feels urgent.
Changes are made carefully by people who are careful, rather than through a process that would catch a mistake.
Uptime looks excellent on paper. Nobody has measured what it costs in unrecorded effort to keep it there.
What this domain examines
Monitoring coverage
Whether monitoring reflects what actually matters to the business, or whatever the tooling produced by default.
Alert quality
Whether alerts are actionable, or whether volume has trained the team to disregard them.
Change control
Whether changes follow a process that would catch an error, or rely on individual diligence.
Redundancy and single points of failure
Whether known weaknesses are documented and quantified, or informally tolerated.
Documentation currency
Whether network documentation matches the network as it exists today.
Escalation and cover
Whether operational continuity depends on specific individuals being available.
Why this goes unnoticed
Competence hides fragility. A capable operations team absorbs faults so smoothly that the faults stop being visible to anyone above them. The work becomes routine, the routine becomes invisible, and the organisation concludes that the network is healthy because nothing has reached it.
That conclusion holds right up until the moment the absorption stops. The trigger is usually a person rather than a system. Someone leaves, someone is on leave, or someone finally stops working around a problem they were never asked to work around. What surfaces then is not a new fault. It is an old one that was being manually suppressed.
This is also the domain where alert fatigue does the most damage. A monitoring platform producing hundreds of low value notifications trains people to dismiss them, which means the one that mattered is dismissed with the same reflex as the rest. Coverage looks complete on a dashboard, while the practical detection capability has quietly dropped to near zero.
How Network Operations is scored
Every domain in the Operational Confidence Index is scored 6 to 18. A higher score means lower operational confidence, so the highest scoring domain in your result is the one to start with.
What we usually hear
“Our uptime figures are excellent.”
They may well be. The index asks what those figures cost to achieve, and whether the effort behind them is recorded anywhere or held by two people who have stopped mentioning it.
“We have monitoring in place.”
Most organisations do. The question is whether anyone acts on what it produces. Coverage and detection are different things, and alert volume is usually what separates them.
“We are a small team. We know our network.”
That is often true, and it is also the risk. Knowledge held by a small number of people without documentation is a dependency, not a control.
“Nothing has gone seriously wrong.”
That is worth knowing. It is not the same as knowing why. The index distinguishes between environments that are resilient and environments that have not yet been tested.
Questions about this domain
How is network operations scored in the OCI?
The Operational Confidence Index uses six questions, each scored 0 to 3, giving a domain range of 6 to 18. A higher score means lower operational confidence, so a high score here indicates this is where to start.
Does this require access to our network?
No. The index is a self assessment answered by your team. Nothing connects to your environment and no data leaves your organisation beyond the answers you choose to give.
Who should answer these questions?
Whoever has the clearest view of daily operations. If two people would answer differently, that difference is itself worth surfacing.
Will this recommend monitoring tools?
No. Rullan Scott holds no manufacturer relationships and receives no commission. The index identifies where the gaps are, not which product to buy.
What happens after I complete it?
You receive a score, a breakdown by domain, and a report. If you want to talk it through, you can book thirty minutes with a cofounder. There is no follow up unless you ask for one.
The other three domains
Find out what your network operations really cost.
Twenty four questions across four domains, five minutes, and a clear view of how much effort is holding your environment together.
Run the OCI Diagnostic