I have recently held an email converstaion with an experienced government auditor who stated that –
Business process improvement methodologies (Lean, Six Sigma) are heavily reliant on metrics / measures.
Unfortunately, the data councils collect for their performance indicators is usually not adequate not robust enough for meaningful improvement initiatives. This is because:
· it is reported to the government and to regulators and this leads to ‘gaming’ behaviours
· it is centrally imposed and does not aid local management
· the data is manipulated and averaged so the ‘voice of the process’ cannot be seen
There is little understanding of the theory of variation and its importance in process improvement. Most councils’ performance compare 2 or 3 points in a chart and conclude that things are getting better or getting worse – this is flawed as you are probably aware.
My response was that during my literature review I came to a conclusion that fits nicely with lean but not well with PI’s, KPI’s etc that disatisfaction was the true ‘measure’. This was because the very personal nature of responses to things on a rikert scale are subjective, as is satisfaction as a whole, but if one can get people to tick a box that they are dissatisfied and explain why they are, you are gaining the ‘voice’ response rather than the ‘exit’ one employed in Hirschman in his development of game theory around East German politics. I’ve no proof that this is true. but the work that rol have done with GovMetric in using a slightly broader brush of satisfied/neither/dissatisfied should provide some guidance (and I am in touch with them, although my council is currently not a user of anything).
My proposal, which is really of no benefit as a government metric, unless one counts the numbers and compare them with the whole, does tend to get around some of the issues mentioned. It can also help assist channel migration in the context that the same reporting factor needs to be used across all channels and identify any particular failures in channels i.e. ‘phone system failures, too complex web site, stroppy reception etc.