Waiting Time in Processes: Why Work Mostly Sits Rather Than Moves

Work sits idle in most processes because tasks wait for decisions, approvals, handovers and free capacity, and hands-on time is often only a small share of lead time. To reduce waiting time, first measure flow efficiency, the share of active work in total lead time, and then act where the queues are longest.

Most executive teams try to shorten lead times by getting people to work faster. The lever usually lies elsewhere: an order can take 25 days to pass through a company and be worked on for only two of them. This article shows how to measure that, why queues grow so sharply at high utilisation, and which measures reduce each type of waiting.

What is flow efficiency and how is it calculated?

Flow efficiency is the share of time in which an order is actively worked on, measured against total lead time from order to delivery. The calculation is simple: active working time divided by lead time. An example with assumed values: a customer order is worked on for two days in total, but 25 days pass between the order and delivery. Flow efficiency is then eight per cent, and for the remaining 92 per cent of the time the order sits in a queue.

Measuring it needs no software. A team takes ten completed orders from recent months, reconstructs every status change from tickets and timestamps, and records for each step whether work was active or waiting. The figure is a baseline for every later measure and belongs to the set of measures described in the article on flow metrics: throughput, work in progress and lead time. Teams are regularly surprised by how low the values are in knowledge work, because each person is busy in their own step and the waiting between steps belongs to nobody.

Why does waiting time grow so sharply at high utilisation?

Waiting time grows disproportionately at high utilisation because each randomly arriving task is more likely to meet an occupied station. Queueing theory describes this for a simple case with one processing station, random arrivals and randomly varying processing times: mean waiting time equals mean processing time times utilisation divided by one minus utilisation.

Real processes deviate from the model, but the direction holds: going from 80 to 95 per cent utilisation multiplies waiting time by almost five. The call for fully utilised departments therefore lengthens lead time systematically. The article on introducing Kanban describes how to steer a flow that is guided by lead time. The link between work in progress and waiting is also set out in the article on decluttering the project portfolio.

What types of waiting time exist?

Six types of waiting time recur in companies, and each has its own remedy. The table sorts them by cause.

Waiting for handovers and alignment grows with the number of teams involved. The article on dependencies between teams describes how to reduce them.

The approach in four steps

You reduce waiting time in four steps that a team can complete in a few weeks.

1. Map the value stream: an order is drawn from order to delivery as a sequence of steps, with processing time and waiting time for each. The method comes from lean production and is called value stream mapping.
2. Identify the bottleneck: the station with the longest queue and highest utilisation determines the lead time of the whole process. The first measure starts there.
3. Assign the cause: each wait is assigned to one of the six types. The remedy follows from the table.
4. Measure the effect: after four to six weeks flow efficiency is determined again. If it rises, the measure stays, otherwise it is adjusted.

At executive level three decisions count: who decides within what time, from what amount an approval is needed, and how much buffer the bottlenecks receive. All three can be changed without rebuilding a process. The effect on delivery time is shown in the article on shortening time to market.

Checklist: does your work sit more than it moves?

Your work mostly sits idle if at least three of the following statements apply.

• You know the lead time of a typical order but not the active processing time.
• The departments are fully utilised, and delivery still takes longer than expected.
• Tasks are passed on in packages or in fixed cycles.
• Approvals need more than two signatures.
• Questions after the start are the norm.

Reducing waiting time with Niehues Consulting

Niehues Consulting is a consultancy for AI management and organisation design, seen from the delivery perspective, based in Düsseldorf. Founder Marin Niehues builds organisations that deliver, in projects ranging from VC-funded scale-ups to DAX 40 corporations and in more than 20 countries on six continents. In the service areas of delivery management and organisation design he clarifies roles and decision paths as well as the interfaces where work gets stuck in practice.

The approach follows the Niehues Delivery Model. For waiting time, two of its test questions matter most: who decides, who delivers, who uses? And which roles does the operation need, and how do decisions and interfaces run? The starting point is a free initial consultation of 30 minutes, which you can arrange via the contact page. Scope and terms of an engagement are discussed there.