Delivery & Steering

Embedding Continuous Improvement: How It Becomes a Routine of Leadership

Marin Niehues17 August 202611 min read

Marin Niehues speaking with a microphone in front of a screen showing a diagram of agile processes.

Continuous improvement stays effective when it is built into the roles and rhythm of daily operations and managers run it themselves. Its Japanese model is kaizen, and in German-speaking companies it is known as KVP, the continuous improvement process. This article explains the terms, shows why improvement programs flatten out after the initial phase and sets out four building blocks with which executive management makes improvement a routine. The yardstick is delivery: improvement counts when lead time and on-time delivery change measurably.

Where the article touches on co-determination and the works council: This article provides an overview and does not constitute legal advice.

What is continuous improvement, and how does it differ from kaizen?

Continuous improvement is the organized, permanent improvement of processes in small steps by the employees themselves. Kaizen is its Japanese model and literally means “change for the better”. It became widely known in the West in 1986 through Masaaki Imai's book “Kaizen”.

The difference lies in ambition. Kaizen describes a mindset that includes every level, from the shop floor to the top team, and forms part of how leadership sees itself. Continuous improvement, or KVP in German usage, usually refers to the procedure that organizes this mindset, with fixed teams and meetings as well as suggestion channels and metrics. A procedure without the leadership mindset turns into an administrative task, and a mindset without procedure remains a mission statement. Effective systems combine both.

A third term belongs to the pair. In lean, kaikaku denotes the big leap, the fundamental redesign of a process. For ongoing improvement, many small changes with fast feedback beat the big overhaul. The big step is needed when the process itself is fundamentally misdesigned.

What is the PDCA cycle?

The PDCA cycle is the working method of all continuous improvement and consists of four steps: plan, do, check, act. It goes back to the statistician Walter Shewhart and became known through W. Edwards Deming, who taught it in Japan in the 1950s.

Deming himself later preferred the term PDSA, with “study” instead of “check”, to stress that the third step is about learning and not merely about control. In practice, what matters is that every cycle ends with a measurement. Without a baseline there is no check, and without a check the cycle becomes a sequence of opinions.

Why improvement programs flatten out after the initial phase

Improvement programs typically start with energy. There is a launch with trained facilitators, a series of workshops and the first visible successes. After a few months the project budget ends, the facilitators return to their functions and the routines peter out. The reason lies in the design: the program was set up as a temporary initiative and never transferred into line management.

Improvement takes root when it becomes part of the work itself. For that it needs the same properties as any other operational process: an owner with a fixed slot in the diary, as well as a metric and the right to take decisions. If one of these is missing, improvement competes with day-to-day business for attention and regularly loses. The underlying method for office and project environments is described in the article Lean management in knowledge work.

How do you embed continuous improvement in a company?

Continuous improvement takes root through four building blocks: ownership, rhythm, metrics and decision rights. Each block closes one of the gaps at which temporary programs fail.

The first building block is ownership. Every value stream receives a process owner from line management who answers for lead time and quality and sees improvement as part of the role. The process owner appears in the organization chart and in the objectives agreement, and their managers ask about it in day-to-day business. A central improvement function supplies methods and coaches teams, while accountability for the result stays with the manager responsible for the process.

The second building block is rhythm. Improvement needs fixed meetings that take place even under heavy workload. In proven systems the team meets daily for fifteen minutes at the board, or at least weekly in knowledge work, and discusses deviations and next steps. At unit level a monthly review of the metrics follows, and executive management checks quarterly which bottlenecks remain open and where decisions are missing.

The third building block is metrics. What is measured is what the customer and operations feel: lead time and on-time delivery, plus the share of rework. Activity figures such as the number of improvement suggestions submitted say little about delivery. Which measures form a robust set is covered in the article Measuring delivery capability.

The fourth building block is decision rights. Teams improve what they own without submitting every change for approval. Leadership sets guardrails for this: within a defined frame for quality and cost the team decides, and beyond it the next level decides within an agreed deadline. A time budget completes the block. Without fixed capacity for improvement, it is cut at the first peak in workload.

Improving at the bottleneck

An improvement system works when it targets the right points. According to the theory of constraints, which Eliyahu Goldratt described in “The Goal” in 1984, one bottleneck in every value stream determines the throughput of the whole. An improvement outside the bottleneck does not raise the throughput of the whole, and if it lies upstream of the bottleneck, it lengthens the queue there. Leadership therefore first clarifies where work is waiting and concentrates improvement capacity there. How to find these points is shown in the article Finding bottlenecks in value creation.

Examples of continuous improvement in the office

In the office, continuous improvement targets idle time and queries as well as rework. Four examples show what a PDCA cycle looks like there.

  • Invoice approval: the team measures how long invoices wait for approval and tests a threshold below which the business unit approves without a second signature.
  • Quotation: a checklist of mandatory details is completed before costing, and the share of quotes issued without queries is measured.
  • Month-end close: the reconciliations with the most corrections are standardized first, and the number of days to close is measured.
  • Recruitment: fixed interview slots in the business units' calendars shorten the time from application to offer.

What the examples have in common is that each change is small, is tied to a measurable value and is assessed after a few weeks.

How do you motivate employees for continuous improvement?

Employees take part in continuous improvement when their suggestions are implemented quickly and they see the effect. The strongest motivator is the experience that an idea is tested and introduced within days. A suggestion that sits in an evaluation loop for months discourages more than any missing reward.

Rewards for improvement suggestions are a supplementary instrument with clear limits. A meta-analysis of 128 experiments showed in 1999 that tangible, performance-contingent rewards can reduce intrinsic motivation for interesting tasks. For employee suggestion schemes this means that rewards suit larger suggestions with measurable savings outside the employee's own area. For daily improvement in one's own process, recognition and a fixed time budget work better. In Germany the principles of a company suggestion scheme are subject to co-determination by the works council, so a reward system is designed together with it.

Typical breaking points in leadership

Three patterns damage improvement systems particularly often. The first is the add-on task: employees are expected to improve without time being freed up for it. The second is a change at the top. A new head of unit brings their own priorities, and the existing routine ends if it is not written into roles and objectives.

The third pattern concerns the metrics themselves. When a metric becomes a target with consequences, teams improve the number instead of the process behind it. The economist Charles Goodhart described this effect in 1975, and it is known as Goodhart's Law. Leadership counters it by using metrics for diagnosis and regularly cross-checking the effect on the customer.

The role of Niehues Consulting

Niehues Consulting sets up improvement systems so that they run without external support after introduction. The approach follows the Niehues Delivery Model with its phases understand, build and introduce. Three test questions are decisive for embedding. In the understand phase: who decides, who delivers, who uses? In the build phase: which roles, decision paths and interfaces does operation need? In the introduce phase: who owns operation from day one?

On this basis the first step is a picture of where work waits and where decision paths cost time. Roles and rhythm are then designed together with the set of metrics and tested in one area that shows the benefit in the real process. Finally, line management takes over the system with enabled managers. Marin Niehues steers the work as delivery manager, project manager or product owner, together with internal or external teams where required. How improvement systems fit into organizational design is described in the article characteristics and metrics of a high-performing organization. The free 30-minute initial consultation is available to assess your starting position.

Frequently asked questions

What is continuous improvement?

Continuous improvement is the permanent improvement of processes in small steps by the employees who run them. It is based on the PDCA cycle, which takes every change from a baseline to a new standard. It becomes effective when it is anchored in the roles and meetings of line management.

What is the difference between continuous improvement and kaizen?

Kaizen is the Japanese mindset of constant improvement at every level, while continuous improvement usually refers to the procedure companies use to organize that mindset. Kaizen emphasizes how leadership sees its role. Continuous improvement describes the teams and meetings in which improvement takes place.

What is the PDCA cycle?

The PDCA cycle is a four-step method: plan, do, check, act. It goes back to Walter Shewhart and became known through W. Edwards Deming. Every cycle ends with a measurement that decides whether the change becomes the standard.

Why do improvement programs fail after a few months?

The most common cause is their design as a temporary initiative. When the budget ends, facilitation and meetings end with it, and nobody in line management has taken ownership.

Which metrics suit continuous improvement?

Suitable measures are those that customers and operations feel directly, such as lead time and on-time delivery. The share of rework belongs in the set as well. They show whether delivery performance changes and can be collected per value stream.

Do rewards for improvement suggestions work?

Rewards work for larger suggestions with measurable savings, while for daily improvement in one's own process, fast implementation and recognition are more effective. Tangible rewards can reduce intrinsic motivation. In Germany, executive management designs the reward system together with the works council.

How long does it take for an improvement system to work?

First effects show in a value stream after a few weeks, because short cycles provide fast feedback. Embedding it in the roles and rhythm of the whole organization takes several quarters in which leadership consistently keeps to the routines.

Let’s talk about your initiative.

In 30 minutes we clarify where your organization stands and which step will pay off most now.

Request a consultation

30 minutes · free · no obligation