Contradictory KPIs. When KPIs should agree, but don’t

Contradictory KPIs are a common problem and can rapidly undermine management confidence in the KPI system. Even someone who knows nothing about measurement knows that if you look at, say, a figure for widget production in July 2007, then you would expect any other figures for widget production in July 2007 to be the same, wouldn’t you? So if they are exposed to similar-looking measures that give contradictory or differing figures, it shakes that person’s faith in the measure and makes it much less likely they will use the information for decision-making. Contradictory KPIs also give the opportunity for people to use this difference to completely reject what the data is saying, particularly if they don’t like the message.

There are a lot of reasons why the situation could arise. Sometimes this comes about because we’re actually talking about two different things with a similar-sounding name. Other times there are variations in how the KPI is calculated. Often there are just mistakes or failure to include certain datasets in the KPI calculation. All three of these types of problem can be eased with a KPI cheat sheet.

Deal with multiple sets of ‘similar’ data

  • Create a KPI definition sheet for each KPI/measure – this makes sure you are talking about precisely the same measure.
  • Track, record and investigate incidents of contradictory values – this helps you track problems and spot patterns in errors.
  • If the discrepancy is a symptom of some underlying structural issues with the data, be open and honest about this and develop a plan for tackling the underlying issues.

Like most problems in life, the loss of confidence from contradictory KPIs gets worse the more you ignore it. Nip it in the bud and you can retain team confidence in your KPIs, but leave it and you will quickly have an unrecoverable situation.