Measures, metrics, key performance indicators (KPIs). Unsure about which term to use? You’re not alone; and in all likelihood, you hear these terms bandied around interchangeably. Try doing a Web search, and you’ll come up with a bunch of definitions that probably won’t clear up matters much. In the same way we misuse grammar, perhaps it doesn’t matter too much if we use these terms incorrectly, so long as we get our message across. I’m sure I’ve been guilty about misusing them in the past, but this is how I correctly think of them.
We measure performance to evaluate and compare how we’re doing. A metric is the discrete value of a measure, for example the number of customers by product. The problem with metrics is that they don’t give us any context to determine whether the value is good or not.
I’ll give you another example. My son attended a soccer camp, and at the beginning they evaluated the kid’s skills – how many times they could dribble the ball around two cones in a figure eight in a minute, how many times they rebound a ball against a wall in a minute, and so on. At the end, his report card said seven for dribbling and twenty-three for rebounds—but that means nothing unless I already have an idea of what is good or bad. If I knew, for example, the national average in his age group was five for dribbling and twenty for rebounds, it would be more helpful.
Additional contextual information along with our metric is referred to as an indicator, or in more fancy terms a key performance indicator (KPI).
Moving from a Metric to a KPI
We move from metric to KPI by first establishing a target from which we can calculate achievement (as a percentage). We also need a scoring system, so we can determine whether we’re doing good, average, or bad. As we accumulate more data over time, we can compare present and past performances – did we improve? We can also start calculating trends. Our targets may go up or down (e.g. costs or accident rates), so is our actual performance improving at the same rate (or better) relative to our new target?
This type of information is what you find in performance management applications. KPIs are linked to strategic objectives and help express execution in quantifiable terms. They provide quick insight into trends and summary information and drilldown on a dimension (e.g. organization or product), so you can pinpoint the cause of performance problems. For example, imagine a global semiconductor company that’s experiencing a high level of product warranty returns. Using a performance management application, they could drill down on this KPI and identify the source of the problem – a supplier shipping unreliable components.
We can group the metrics upon which our KPIs are derived into three categories:
- Input metrics measure what you did to get the desired result. In our soccer example, these would include how much we spent acquiring players, number of players, depth in position, number of practices, and so on.
- Output metrics measure what the results were. This will give us a sense of what our games were like: number of corners, fouls, free kicks, goal kicks, goals conceded, goals scored, off-sides, passes, percentage in opponent’s half, percentage possession, saves, shots, and tackles.
- Outcome metrics (the most important) measure whether or not we achieved our goal. Did we win the game, the league, the cup?
While it’s important to have a mix of input and output metrics, these are more likely to be associated with operational systems and appear on dashboards. Our scorecards need to have outcome KPIs and show progress over time.
Here’s a real world example from a public health and safety organization that had a goal of reducing the occurrences of a disease in their community by increasing the number of immunizations being given. To this end, they started promoting free immunizations on their website and other mediums. Their KPIs included:
- Dollars spent on immunization services
- Information downloads on immunization and where to get it
- Number of immunizations
- Number of occurrences of the disease in the community
Which of the above is the outcome KPI that most matters?
In part four of this series, I’ll cover more KPI characteristics, best practices, and other considerations.