ChatDeskOps

ChatDeskOps  /  Coverage arithmetic

Section 02

One live seat costs about 4.9 people

This page exists because mis-pricing coverage is the single most common way a partner loses money on a 24/7 process. Read it before the commercials.

Hours in a year that must be covered365 × 248,760
Working hours one full-time person delivers52 wks × 40 hrs2,080
People per continuously live seat, before shrinkage8,760 ÷ 2,0804.21
Shrinkage — leave, sickness, training, attrition× 1.16+16%
Headcount per continuously live seat 4.21 × 1.16≈ 4.9

A seat that must be live every hour of the year needs 8,760 hours of attendance. One full-time person delivers about 2,080. That is 4.21 people before anyone takes a day off.

Shrinkage — annual leave, sickness, training days, coaching, and the gap between one person resigning and their replacement being productive — adds about sixteen per cent. That is where 4.9 comes from, and it is a floor rather than a comfortable estimate.

Twelve seats is roughly sixty people. Sixty salaries, sixty transport seats at night, sixty background checks, and a bench underneath. Partners who quote this process off the seat number rather than the headcount number discover the gap in month two, and by then the roster is already committed.

It is also why the seat band starts at twelve rather than lower. Below twelve you cannot build a roster that covers every hour without giving someone a pattern they will leave within a quarter.

The bands

Four coverage bands, and what each one demands

Client customers sit across three continents. The roster is built to the shares below and reviewed monthly against actual arrival patterns — but the shape does not change much, because customers do not change time zones.

Australia and Asia-Pacific22% of roster
22%
United Kingdom and Europe30% of roster
30%
US eastern and central32% of roster
32%
US west and overnight16% of roster
16%

Twenty-four hours of client time, and the share of the roster each band absorbs. Every one of them has to be staffed on a Sunday at four in the morning to the same standard as a Tuesday at eleven — because a customer with a production issue at three a.m. does not accept that the desk is thin overnight.

BandShare of rosterCharacter of the trafficStaffing note
Australia and Asia-Pacific~22%Steady business-hours volume, high how-to shareThe most comfortable band to staff from India
United Kingdom and Europe~30%Heaviest chat concurrency; morning peak is sharpNeeds the deepest bench for the peak hour
US eastern and central~32%Largest single band; broad contact mixThe band where escalation volume concentrates
US west and overnight~16%Lower volume, higher severity — production issues surface hereLowest volume, highest cost per contact to staff

The quietest band is the most expensive one, and it is the one a struggling provider is most tempted to thin.

The credit

What an uncovered hour costs

Coverage is the only measure on this process expressed as one hundred per cent, and the only one with a financial consequence attached directly to it.

TermBasis
What counts as an uncovered hourAny clock hour in which the agreed minimum staffing for the band is not met for a continuous period of 15 minutes or more, measured from the platform's own login and presence data.
The creditOne seat-day is credited for each uncovered hour.
The capCoverage credits in a month are capped at 5% of that month's invoice.
What is excludedClient-side platform outage, and force majeure events notified at the time and evidenced. Provider-side power, connectivity, transport and staffing failures are never excluded.
How it is evidencedFrom the client platform's presence and login data, not from the provider's roster document. The roster is the plan; the platform is the record.
Where it is reportedIn the monthly coverage report submitted with the invoice, whether or not any credit arose.
The cap protects both sides. It stops a single bad night becoming an existential invoice, and it stops the credit being treated as a cheap way to buy out of a coverage obligation. Persistent coverage failure is handled through the escalation ladder and the remediation plan, not through credits.
Holding it

What actually keeps coverage at one hundred per cent

The controllable levers

  • A workforce-management analyst who owns the roster as a full-time job, not a spreadsheet a team leader updates on Fridays.
  • A bench sized against actual attrition rather than planned attrition.
  • Team leadership present on the overnight band — a night shift without a team leader is an uncovered shift waiting to happen.
  • Schedule adherence measured daily at 95% or above, because coverage fails through lateness long before it fails through absence.

What breaks it

  • Night transport that is unreliable in monsoon or on public holidays.
  • A shift pattern that rotates too fast for people to sleep, which produces attrition that looks like bad luck and is not.
  • Treating the overnight band as the place to put new agents because it is quiet.
  • Carrying a vacancy for three weeks because the seat still bills for the first three days.
The overnight band is not a training ground. It has the lowest volume and the highest severity — production issues surface at three in the morning. Staffing it with the newest agents is the most expensive false economy on this process.

Next step

If the arithmetic on this page changes your pricing, that is the point.

Better to find it now than in month two of a twelve-month term.

See the commercial terms Read the provider economics