Long queues are visible, frustrating and expensive. They are not, however, automatic proof that an organisation needs more people. In banks, clinics, contact centres, public services and field-support operations, waiting time often emerges from a combination of demand peaks, uneven task times, rework, approvals and poorly positioned skills. Treating every queue as a staffing shortage can add cost without improving flow.
A queue is not automatically a staffing problem
Industrial engineering starts by separating demand, capacity and variability. Average workload can look manageable while customers still wait because arrivals cluster around paydays, outages or reporting deadlines. Capacity may also be trapped in non-value-adding work: agents searching for information, supervisors authorising routine exceptions, or cases returning because required data was missing. The system then behaves as if it is understaffed even when paid hours appear sufficient.
African operating conditions can amplify this effect. Connectivity interruptions, transport delays, power instability and scarce specialist skills create variability that a monthly staffing ratio will miss. The practical question is therefore not “How many people do we have?” but “How much reliable completion capacity is available when demand arrives?”
Use a simple operating test
Before adding headcount, review one representative week and test four questions:
- Arrival pattern: When does work enter, and how sharply does demand peak?
- Service time: How long does each case really take, including hand-offs and corrections?
- First-time quality: What proportion leaves the process without returning?
- Effective capacity: How much available time is actually spent completing customer demand?
Track these measures by hour or shift, not only as monthly averages. A small time-based view often exposes the real leverage point: staggered shifts, simpler routing, better information at entry, fewer approvals or a protected specialist window.
Manage variability before buying capacity
The goal is not maximum utilisation. Running every person at 100% leaves no room for variation and makes queues grow quickly. Instead, define a service target, protect capacity at predictable peaks and remove avoidable demand. The daily-management system should show arrivals, completions, backlog age and first-time-right performance together. If yesterday’s fixes keep returning, use a systems-thinking review to trace the feedback loops creating repeat demand.
Turn the signal into a decision rhythm
Run a weekly capacity review with operations, customer service and enabling teams. Agree which causes will be removed, which peaks require flexible cover and which service promises need redesign. Add permanent capacity only when measured demand remains above effective capacity after flow losses have been addressed. That sequence protects service levels, scarce skills and margins—while ensuring that investment improves the whole operation rather than hiding a poorly designed process.