In my last blog I wrote about the natural way of working in process, now I will follow up and writ about how you can measure the process.
Time, money and units. That’s it. In theory this is simple. You want to measure the effectiveness, cost or contribution of a single event or a series of events and you choose one of the three. Or all of them.
At the same time it is really difficult to collect data and format it so that you can present something that makes sense. If the work in a process is scattered across multiple locations across the globe and the work is split on numerous items, you will need to find some shortcuts.
- Tip1 Don’t go bananas, find a couple of areas and add data gradually and find the data at the lowest possible level, like sales of item X in store 123.
- Tip2 Keep all the raw data in one place and find ways of trimming it, putting this information in a secondary place, to ease analysis (keep a summery log of what you are trimming to continuously check that you have made the right assumptions).
- Tip3 Top level numbers are not always the sum of the lower levels; make sure you know when they are not.
- Tip4 KPI’s are important from a management point of view, but OPI’s can be more helpful in the short run and help make the right operational decisions, find a way to balance both with the time you have on hand. In the best of worlds one will support the other.
- Tip5 Try some versions before settling for what will be your standard.
So, related to tip5, what measurements should the raw data support? We need to know how effectiveness, cost or contribution is spread across the process (steps A to G in the figures bellow). We need to be able to see the relationships between the processes and we need to see development over time
We also need to know the change between two points in time; or different periods can be summarized in an index change. The index number is easy to understand as anything under 100 is a lowered result and anything over is an improved result.
The value you are looking at can be measured as an average, you can look at max and min values, the standard deviation or the accumulated value. The advantage of one number is of course simplicity in in communication. This number can also be powerful if shown over time to see development and trends.
We used a lot of index numbers in IKEA and it made communication a lot easier. The most popular index was last year’s result vs. this year. In some cases it was irrelevant but in cases like accumulate year result vs. accumulated current year result it was crystal clear if you had a tough road ahead to reach you goals or not. What we started to improve was to communicate the index between planned and executed. This is a crucial index, but was not used much. For me it says a lot about the business.
It is also important to find a way of measuring the change that occurs in a process step (IN vs. OUT) in order to find the contribution of the step or to see if there is a buffer effect; i.e. that there is an over capacity to be able to catch up in this step or not. This can be seen as a good thing or you can see it as just having too many resources in this step. In the same way you can see if one step continuously is lagging, then you might need to add resources here. One thing that can be devastating is variance in a process step. This is the level of unpredictability and needs to be managed either by building in a buffer in the process or as safety stock, either way, it is what costs money.
The thing is, if you know what you want to measure you need to get started, NOW, don’t sitt around. Because, unless you start measuring it, you’ll never be able to analyze it. You might have to rethink how you measure, but sitting around thinking about it will get you nowhere. Waiting for perfection is just another way of failing.
So now, off you go! Get it done already! You’ll be happier a week, a month or possibly 6 months from now, depending on what you decide to do…