The art of making things small

There are different ways to improve processes. 'The art of making things small' refers to the choice to reach your goals in small steps. This relates to principle 14 of the Toyota Way: "Become a learning organization through relentless reflection (hansei) and continuous improvement (kaizen)". Why would you want to make things small? Because a pitfall of projects is that we discuss for weeks or even months before anything changes. The  risks with this approach are:

• The longer it takes for the first change to take shape, the more people who experience the problem loose faith that it will be resolved. The support for any change crumbles.
• Many of those who perform the work in the process where the problem occurs are not involved in the thinking process. This means that you need good communication (which is often problematic) and often the much feared 'resistance to change' starts to take shape.
• The solutions are developed separate from the work floor, separate from the (physical) place where the problem occurs. That means there's a real risk that the solution will not exactly fit the complexity of the situation where it's supposed to work.
• Perhaps the solution will be piloted, but by now those who are in favor of the solution have argued for weeks or months to promote it and they now have to prove that it works.They have connected their personal reputation to the solution and will not like to see that it fails. Those who are not in favor of the solution hope that it will not work, they might even influence the results of the test to make it not work. Will this be a good test?

Recognizable? I have seen this happen over and over again. The art of making things small is a method that approaches it very differently. It is based on the teachings of Roger Resar on Reliability. The essence is simple: reduce changes to a size that you can literally test tomorrow. A team can use the next steps do achieve this:

1. Define the process where change is needed.
   For example: it takes too long before general practitioners receive the letters form the specialist about the results of patients.
2. Describe the process in four or five steps.
   For example: appointment - dictate letter - write letter - send letter.
3. Define where a problem occurs. If necessary: describe this step in four or fives steps.
   For example: time between appointment and dictating takes too long.
4. Define a change that you can test tomorrow. Choose the most easy circumstances for the test.
   For example: one specialist will dictate one letter for one patient right after the last appointment on the less busiest day of the week.
5. Evaluate whether the test worked. Just yes or no. Evaluate why it did or did not work. Define the next test.
6. Keep on testing the idea or new ideas until a good countermeasure has proven itself in practice. Make this the standard new way of working.

To solve the complete problem of the total throughput time, the time between the appointment and dictating is probably only part of the problem. There are likely to be multiple causes. The idea is that more then one test run simultaneously. If a team is experienced, four tests can run at the same time, each test lasting one day or week.


Except addressing the risks as described above, there are other reasons to apply this method:

• By making changes smaller, they are pulled into the sphere of influence of the team. They can achieve more changes themselves.
• There will be more learning. Learning by doing is more effective.
• Decisions are more based on facts, more based on experience. Opinions and emotion become less influential. 
• Changing becomes more fun, because the people that undergo the change decide and evaluate the change themselves. There is more ownership, more pride.
• Less time is spent on meetings.

This method is based on the 'plan-do-check-act' cycle of Deming or perhaps more accurately from Shewart. Reread the six steps above to recognize them.

There are also pitfalls with applying the art of making things small:

• Not enough analysis why there is a problem
  Suggestion: you do need an understanding of the cause of the problem. Often this is clear enough and you can start testing directly. But sometimes you do need further analysis to assure that you are not addressing aspects that are not relevant to the problem. 
• Not enough coordination of the tests. The team goes testing changes without keeping track of the effect on the problem that needs to be solved.
  Suggestion: keep going through the complete cycle (see above). Not only for each test, but for the complete problem. Someone is responsible for the coordination (the owner of the problem).
• Reluctance to start testing because the person that does the test is not convinced it will work as the new method of working.
  In the example the specialist might not want to test dictating directly after the session for one patient, because she expects it might work for one patient per session for one test, but not for all patients every day.
  Suggestion: test anyway! Doing a test is no obligation to keep doing it, also not if it works. You can only judge whether the idea is feasible after the test. Then you can decide based on facts instead of imagination. More importantly, doing the test can lead to a better idea that is feasible.

Achieve great things by making them small!