The 80/20 target is used almost everywhere, yet few know where it comes from. What service level tells you – and what it hides.
Service level is not a target. It is a trade-off.
The common phrasing "80/20" means 80 per cent of calls answered within 20 seconds. The numbers look like an industry standard, but they are historical rather than derived from any requirement. For technical support handling complex cases, 80/60 may be more sensible than 80/20 – for an emergency line, far too slow.
Why the average is not enough
An average wait of 25 seconds sounds acceptable. It can mean almost everyone gets through immediately while some wait ten minutes. The average hides exactly the cases that generate complaints.
That is why service level is expressed as a share within a threshold rather than a mean. It is worth pairing it with the longest wait in the interval – that shows how bad the worst case actually was.
The non-linear part
Wait time does not scale proportionally with staffing. As utilisation approaches capacity, wait time rises disproportionately: the last few percentage points of availability cost considerably more than the first. Conversely, one extra seat on a thinly staffed shift often yields an unexpectedly large improvement.
Practical consequence: planning in short intervals – usually 15 or 30 minutes – produces usable results; daily averages do not. A day can meet the target and still contain two hours in which nobody got through.
Abandon rate as a corrective
The abandon rate – the share of callers who hang up before being answered – always belongs next to it. A met service level alongside a high abandon rate usually means the difficult calls gave up, and the metric benefited.
What should set the target
Not the industry average, but the expectation attached to the channel. People who call expect an answer in seconds. People who write accept hours – provided the answer is then complete. A single target across all channels optimises for none of them.
