Applied6 min readLesson 1 of 5
What production support actually is
Production support keeps the business running on top of systems written by people who left years ago. It is part detective work, part risk judgement, and it rewards method over cleverness.
A realistic week
| Activity | Roughly |
|---|---|
| Investigating and fixing overnight batch failures | 30% |
| Answering questions about how something works | 20% |
| Small changes and data fixes, through change control | 20% |
| Monitoring, housekeeping and capacity checks | 15% |
| Improving things so tonight's problem does not recur | 15% |
That last row is the one that distinguishes a good team from an exhausted one. Every recurring failure is either a fix waiting to be made or an automation waiting to be written.
The skills that matter, in order
- Reading job output accurately. Most incidents are solved by reading carefully rather than by knowing something obscure.
- Knowing the business flow. Understanding what the job is for tells you whether a failure is urgent, and what a safe fix looks like.
- Judgement about restart. Knowing when it is safe to rerun is the single highest-value skill in the role.
- Clear communication. Saying precisely what is broken, what the impact is and what you are doing about it, to people who are not technical.
- Discipline under pressure. Following the procedure at 3am when improvising feels faster.
How the work reaches you
Automated alert
A job abendedA threshold breachedPaged or ticketed automatically
Operator call
Something looks wrongA message needs a decisionOften out of hours
Business report
A report is missingA figure looks wrongUsually the most urgent
Common mistakes
Trying to understand the whole application before acting
Understand the failing step and its immediate context. Full understanding comes over months.
Fixing without recording
An undocumented fix is a problem you will solve again. Write it into the runbook the same day.
Treating every alert as an emergency
Impact drives urgency. A failed report that nobody reads until Monday is not a 3am problem.
What you will see at work
- There will be a runbook per application, of wildly varying quality. Improving one is the fastest way to become valuable.
- Handover between shifts matters enormously. Write it as though you will not be available to explain it.
- The on-call rota is real. Ask how it works, how often, and what support exists, before accepting the role.
Key terms
Check your understanding.
Take this lesson's quiz and save your progress. Free.