It is common to have interesting but inconclusive data. If this happens at an early stage of a project, then that’s OK. Sometimes you do need to wander around the data just to get a feel for it. If this occurs at the end of a project, then you’ve got a problem. Leaders won’t be happy having funded a project that is merely interesting.
Before we get into how to deal with inconclusive findings, let’s be clear that findings are never fully conclusive. If you pretend that they are, then you are misleading your audience. What evidence-based findings do is decrease the uncertainty. Should you open an office in Hanoi or Ho Chi Minh City? You can never know for sure which is best, but if you’ve analyzed the data you can be a lot more confident in making a recommendation.
When you do have data that is so inconclusive that it’s merely interesting, then that’s usually because you’ve not clearly defined the problem you are trying to solve.
Take the choice of Hanoi vs. Ho Chi Minh City. Suppose the company is committed to opening an office in Vietnam and has narrowed it down to those two cities. All things considered, then, the data is bound to lean in one direction or the other. When an analytics project’s goal is to support a specific decision, then it’s unlikely that you’ll end up with a totally inconclusive finding. Similarly, if you have a clear answerable question or hypothesis, it’s likely the evidence will lean one way or another.
Article Continues Below
Contingent Workforce Strategy Survey With ERE and Aptitude Research
If your company currently leverages contingent workers, please share your views in our brief survey.
If you do hit the “inconclusive data” problem, then you need to go back to the stage of getting clarity on what you are trying to solve for. This seems straightforward, but it’s hard enough that it is often a pit of despair for those new to analytics. Persist in asking about the business issue, the options, the decision you’re trying to make, the potential commitment to action, the hypotheses, etc.
When all this happens early in the project, it’s not a problem. If it is at the end of the project, then you need to admit the problem and show that you have incredible clarity on what additional work is necessary to bring the project to successful fruition.