Continuous Improvement on a Small Scale: The Weekly Improvement Suggestion

Why good ideas fizzle out in a company

Almost everyone in a company knows what holds them up at work. The template that has to be adjusted by hand every time. The folder nobody can find. The follow-up question needed for every third order. These things are rarely said out loud, and when they are, it is in passing in the kitchen. There they stay.

Many companies therefore have a suggestion box, a list on the shared drive, or an item on the agenda. After a few months the box is empty. That is usually not because of a lack of ideas, but because nothing happened with the ideas.

This article asks: how do isolated ideas become a regular, small improvement that actually gets implemented? To answer, we first clarify the terms, present the PDCA cycle as a thinking model, examine what studies show about suggestions and improvement, weigh two forms of suggestion systems, and finally describe the weekly improvement suggestion.

A suggestion nobody answers is the last one that employee will make.

What does continuous improvement mean?

Continuous improvement is the ongoing effort to make workflows, services, and products better in small steps. Its hallmarks are small steps, high frequency, and short cycles. The idea behind it: many small improvements add up to a large difference over time. In the West, the concept became known mainly through the book Kaizen by Masaaki Imai (1986). The Japanese word roughly means change for the better.

An improvement suggestion is a concrete idea from an employee for how a workflow at their workplace can become simpler, faster, or less error-prone. It differs from an improvement project, which reorganizes an entire workflow over several weeks, as described in the article on process optimization step by step. The two complement each other: the project reorganizes the workflow, the suggestions keep it moving afterwards.

The effect can be compared with compound interest. An improvement of one percent goes unnoticed. Fifty such improvements a year change a workflow noticeably, because each builds on the previous one. The reverse is also true: small deteriorations that nobody addresses add up as well.

The thinking model: the PDCA cycle

The PDCA cycle goes back to the statistician Walter Shewhart and was made known by W. Edwards Deming. It describes improvement as a loop of four steps. Its value lies in treating every idea like a small experiment, not like a decision that has to be defended.

1. Plan

   Describe the problem in one sentence, the suspected cause, and the change you want to try. Define how you will notice whether things got better.

2. Do

   Implement the change on a small scale, for one week, one workplace, or one type of order.

3. Check

   Compare the result with what you expected. Has the number or the observed problem changed?

4. Act

   If it works, the change becomes the new standard, for example in a template or checklist. If not, it is discarded, and the next attempt begins.

The fourth step is skipped most often. A change that exists only in one person's head disappears with the next vacation. Only when it is written into a template, checklist, or work instruction does it become a standard. How to phrase lessons so that they take effect is described in the article on project closure and lessons learned.

What studies show about suggestions and improvement

Kang, Li, and Kang (2025) examined how employee suggestions influence continuous improvement. They used data from the international research network High Performance Manufacturing covering 308 manufacturing plants in machinery, electronics, and transportation in 15 countries and regions. What was measured was not whether a suggestion box exists, but whether the plant actually implements suggestions and gives feedback to those who made them.

Three results are revealing for small companies. First, suggestions improved continuous improvement directly. Second, part of the effect ran through problem-solving in small groups: where suggestions were discussed and worked on in small teams, they had a stronger effect. Third, strongly centralized decision-making weakened this relationship. Where everything is decided at the top, suggestions reach joint problem-solving less often. The authors stress that their data come from manufacturing plants and were collected at one point in time.

Ikica and Vrtodušić Hrgović (2026) surveyed 244 managers in medium-sized and large hotel companies. The degree to which improvement was actually implemented had a strong influence on results. Employee engagement and data analysis worked through this level of implementation. Quality training, by contrast, had no measurable effect. This fits the finding by Kang and colleagues: what matters less is what people know about improvement, and more whether suggestions are worked on regularly.

Bonus or habit? Two forms of suggestion systems

Two forms have emerged in practice. The classic suggestion scheme collects ideas, has them evaluated by one office or a committee, and rewards the best with a bonus, often depending on the savings. The Japanese form, often called kaizen teian (Salvadorinho et al. 2024), relies on many small suggestions that the person who made them usually implements. Recognition is small, frequency is high.

In favor of the classic model: it makes large savings visible and rewards them. Against it: long evaluation paths delay feedback, and a bonus based on savings makes small ideas unattractive when they cannot be expressed in euros. For small companies I consider the second form the better fit. It needs no committee, and it matches the finding by Kang and colleagues: suggestions work when they are dealt with in a small circle rather than administered centrally. Bonuses are not ruled out, but they should be the exception for big ideas, not the driver for all.

The weekly improvement suggestion: five rules

The weekly improvement suggestion is the tool of this article. Every team contributes at least one suggestion per week. It is decided in a few minutes in the weekly meeting that takes place anyway. For this to work, five rules apply.

• Small: the suggestion can be implemented in a day at most and costs little or nothing.
• In your own area: it concerns your own work or the direct handoff, not other people's work.
• Decided within a week: yes, no, or try it, but never postponed without a date.
• Implemented by the person who suggested it: whoever suggests implements, with support if needed.
• Result visible: implemented suggestions are on a board or list that everyone can see.

The weekly meeting is the right place because it takes place anyway. A fixed five-minute item is enough, as provided for in the article on weekly team planning. If a suggestion affects another area's work, it goes to the person responsible there, with the same deadline. Who may decide what is covered in the article on setting decision rights.

What follows in general

Continuous improvement is less a method than leadership's attitude toward feedback. Whoever reviews suggestions and answers them promptly shows that employees' knowledge counts. Whoever lets them sit shows the opposite, even without saying so. That is why the response time to a suggestion is the most important number of the whole procedure.

This also shows where the limits lie. Small suggestions improve a workflow that is basically sound. If the workflow itself is built wrong, it takes a project. And if suggestions keep addressing the same mistakes, it is worth looking at the causes, as described in the article on errors as process signals. Where ideas can come from is shown by the waste walk from the article on lean in the office.

Conclusion: how do ideas become regular improvement?

The question was
isolated ideas become regular, implemented improvement. The answer lies less in the number of ideas than in how they are handled. The studies show that suggestions work when they are implemented and answered, when small groups work on them, and when decisions are not all concentrated at the top. The PDCA cycle gives every suggestion the form of a small experiment, and the weekly rhythm ensures that none of them is left lying. A company that makes one small thing better every week needs the big improvement project less often.

What you can do this week

1. Add a fixed five-minute item for suggestions to the next weekly meeting.
2. Bring a small suggestion of your own, filled out on the suggestion card.
3. Decide every suggestion in the meeting itself: yes, no, or try it until a set date.
4. Create a visible list of implemented suggestions.
5. After four weeks, check how many suggestions came in and how long the answer took on average.

After a quarter, the list shows whether the habit holds. More important than the number of suggestions is that none went unanswered.