What is triage?

Why it matters

Triage is the highest-leverage step in a support process because it is cheap and it determines everything after. A wrong triage decision costs the full resolution twice. It is also the step most often done by the most expensive people available, which is exactly backwards.

What to know

• Triage should take seconds. If it takes minutes, the criteria are too complex or the data is missing.
• Priority and severity are different: severity is how broken it is, priority is what you do about it first.
• Automated triage on clear cases with human review of ambiguous ones is the pattern that scales.
• Consistency matters more than perfection. Two triagers disagreeing is worse than both being slightly wrong.
• Track downstream re-triage rate, which is the honest measure of triage quality.

An example

Example: a request mentioning cannot log in from an enterprise account is set to high priority and sent to the access queue in under a second by rule. A vague request saying it is broken goes to a human triage queue for a ten-second read.

Common mistakes

• Having senior agents triage everything by hand
• Conflating severity and priority in the same field
• Writing triage criteria that require judgement on every ticket
• Never measuring how often triage decisions get reversed

Questions