Plan – Based on data recieved from marketing and other efforts to understand a particular “thing”, decide what it is you want to accomplish and devise a plan to make it happen. I say thing, because you could be analyzing a customer base for areas of improvement or you could be analyzing a process for improvement. Really this “thing” is anything that can be measured that you may want improved.
Do – Once you figure out what it is you want to accomplish (a goal), and how you’re going to go about accomplishing it, you DO! Implement your plan, be disciplined about following it.
Check – Some people also refer to this step as study. The point is to step back and reflect on the results of your doing. Use the data presented to evaluate the progress, if any.
Act – If the results were good, adopt the change into your process so that there will be no backslide of progress. The purpose of this cycle is continuous improvement and if the cycle did not produce positive results, then it would make senes to abandon the plan and try another avenue for improvement.
The backbone of the PDCA cycle is DATA. Data needs to support the decision making and the goal needs to be measurable. Without data, you are just shooting into the dark. Data will help you prioritize policies. A business can look at the data and say, “well if I do this I’ll see a 50% reduction in production cycle time. If I do this other thing I will see a 35% reduction in production cycle time. I guess I’ll do the first idea rather than the latter”. Of course this is difficult to do, and pareto charts, control charts and multiple regression analysis can help identify areas the copmany would benefit most from improvement in.
A common question that arises from this chart is whether or not the time spent on each task is even or not. In a “lean” type system the plan stage takes up the most amount of time. Planning is a function of management, and often times this planning will require a horizontal integration of different areas within a company. Multiple inputs contribute to the complexity of the planning task, which extends its time frame. In addition, all over the other stages of the cycle are dependent on the first, so it is important to understand the problem and provide a detailed solution. For these reasons planning may take up 50% or more of the cycle time, with the “do” portion being the smallest of the four at perhaps 10%.
There is a large amount of things that come into play around this cycle but I’ll just leave it at what it is, a model for continuous quality improvement. I wanted to note that in many business environment, success is paramount. People want results and often times management is rewarded on results. The goal of this cycle is to make a small incremental improvement NOT a breakthrough type improvement. In breakthrough improvement, often large sums of money are required, while Kaizen should not require large sums of money. I think the companies that reward on the basis of improvement % without looking at overall capital cost are missing the point.
Lastly, I want to reiterate that PDCA is a cycle, and even if a plan is implemented and fails to meet the goals set out, that does not mean that the time spent was useless. It is still important to learn from the cycle and the data that was presented and collected during it. This data can be used to jumpstart the next round.
Here’s a follow up post on application of PDCA to personal goals and growth.