Charts can do a whole lot of things. It’s easy to say “it’s science” or “we’re data-driven” because there’s a chart present, but without understanding where the data came from or how it’s counted it’s pretty hard to know for sure what the chart says: other than that particular measure has increased or decreased over the allotted period of time.
Take kids’ messy bedrooms. If the standard measure was the number of clothes on the floor, there are a few ways to graph and interpret the results. For example, we could record the change in the number of clothes on the floor over time.
Whether you’re using a bar or a line chart, it’s pretty easy to visualize an increase in clothes over time. What’s more difficult is interpreting what it means. An optimist could use the chart as evidence that the child has been more active: more clothes signifies more activity, which represents healthy growing children. Alternatively, the same chart could be used to represent an increase in laziness over time.
Therefore, when trying to fix the problem, what’s the best way to know if our actions are really working? The first step has nothing to do with the solution:
Agree on the measures
It doesn’t matter if you’re using the first or second interpretation above, or something else altogether. What matters is that those involved agree on what each measure means. So the folks you’re reporting to know might believe that for your purposes more clothes on the ground means more laziness. Additionally, your kids might know that you’re tracking their activity, and that’s what you think it means.
This has a significant impact on your overall performance because by setting clear ground rules beforehand, you’re eliminating debate once the work or experiment is completed.
Agree on the time period for measurement
Setting a firm deadline for evaluation helps to align focus across the board. Everyone involved knows what to expect and when. This way there’s no surprise when a serious evaluation comes around.
More often than not, organizations agree on measures but fail to set a clear time period for measurement, so the results are disregarded by pointing out that ‘we didn’t have enough time.’ That’s something that can be easily discussed beforehand, so the measurement period can be adjusted appropriately.
As excited as we get about data, the context is as much, if not more important than the figures themselves. This is why leading organizations spend millions of dollars adjusting and standardizing their measures before making big decisions. It’s critical to know what the numbers really mean before giving anything significant a real shot.
That way, when all is said and done there’s no doubt about whether or not what you’re testing for is really working.