The Psychology of Quality and More
When gathering and analyzing survey data, for example in a customer survey or maybe an internal employee survey, a common approach is to get them to score you in a number key areas that you have selected. For example, many surveys look something like this:
Please circle one number in each line
Widgets work well for me Disagree [ 1 2 3 4 5 6 7 8 9 ] Agree
Spridgets are wonderful Disagree [ 1 2 3 4 5 6 7 8 9 ] Agree
A whole host of questions thus allow you to find out how well you are performing in a number of different areas. You then can use this to focus your improvement efforts in the low-scoring areas.
This is all very well as far as it goes, but it does not address a second and very important question: do your customers really care about this? You could spend a lot of time improving Widgets, perhaps rather than improving Spridgets, yet malfunctioning Spridgets are far more important to customers than a Widget that does not work as well as it should.
The simple trick to resolve this situation is to ask customers about both importance and performance. For example:
Please score both Importance and Performance from 0 (low) to 9 (high)
Widgets Importance [ ] Performance [ ]
Spridgets Importance [ ] Performance [ ]
The next stage is to find a useful way of combining these in a way that eases communication and decision-making. A common approach with two factors like this is to multiply them (for example as is often done in FMEA with risk probability and impact). An approach when exploring the relationship between two factors that is more useful is to plot them on an X-Y chart, as in the Importance-Performance Grid below.
This diagram now clearly lets you prioritise your improvement action, as you start with those items in the bottom right of the grid, where there is high importance and low performance.
The grid also lets you make other decisions, for example you may drop those things which are low importance, and particularly if they are low performance (do we really need to keep making Widgets?). The grid also highlights your high importance-high performance stars. With these, you can make decisions such as to add focus to protect their performance, or even to increase the price you charge for them!
Another use of a grid like this is to show clusters and patterns of correlation, for example where are group of products are all high importance and only moderate performance. A close examination of these might find a root cause of the lower performance, for example using similar standards or designed by the same people.
A trap with such diagrams is to over-simplify them. For example the dotted lines in the diagram could be used to create four simple groups for which four strategies are applied. It is always worth remembering that grids and simple grouping can help give indications, but the final decision of what should be done is always human.
Next time: Concept Screening
This article first appeared in Quality World, the journal of the Institute for Quality Assurance
And the big