A Data Collection Plan ensures that everyone is on the same page with regards to the data plan. It also ensures that stakeholders know what is expected of them. The purpose of the plan is to make sure that the data collected are meaningful and valid and that all relevant data are collected.
A Data Collection Plan helps in being efficient by not wasting resources on collecting data that are irrelevant to the project or not usable. It also helps in presenting a clear path forward by pooling knowledge and presenting it in a clear way for everyone. Not all data may be available or be a perfect fit, with this plan you can bridge the gaps.
You typically make a data collection plan in the measure phase of DMAIC. And you keep it updated from there.
How to make a data collection plan
You always need to work together on this. Get everyone in the same room and fill out the plan together. This way you will get the advantage of all the combined knowledge and reduce the chance you are missing anything.
In the first columns you need to make a link to the process you are trying to improve. You can do this by using the process steps in the SIPOC or in the value stream map. I personally prefer to use the SIPOC for this. The following things are important to add:
- The name of the process step in the SIPOC
- The critical input for that step (without this input the process step cannot start)
- The critical output for that step (without this output the next process step cannot start)
You can add the critical to quality variables for input and output here. Sometimes these are self evident and i leave them out. Sometimes there is discussion among stakeholders and than i put them in the data collection plan.
The next columns in your Excel sheet should be about what you are measuring in each process step. Define what you need to measure by looking at:
- Name the metric you need to measure to reach the stated goals of your DMAIC
- State the data type (discrete, continuous, attribute etc.)
- Operationalize the metric. What is being measured by whom/what? Make it as SMART as possible and check whether all stakeholders are on the same page. See if there are several ways to explain the definition. If there is finetune it until there is only one possible definition left.
- Plan how, when and where to collect the data needed for the metric specified. How can be by using systems, but is can also mean measuring everything yourself.
- Look at the sample size of your data collection. Will it be 100%? Do you need to make a selection? If so how will you do this?
- Determine the frequency of data collection. This can be daily. Any frequency can be helpful as long as it helps you get enough relevant data.
- Determine who is the owner of the data and who will be the owner of the metric.
Sometimes upper and lower specification limits are defined as well. I tend to skip this when there is no data available yet.
Your data collection efforts should lead to new standards to be able to sustain the gains made during your DMAIC. To do this you need to define the final two columns of your excel sheet like this:
- A reporting format and owner
- A standard operating procedure to keep measuring
Below you can find an explanation on different data types to help you get started with your own data collection plan! Knowing your data types is crucial in having an effective data collection plan.