Data-Based or Experienced-Based Decision Making: Which Is Better?
February 05, 2021
According to The Economist, productivity rates for American builders have plunged by half since the 1960s. This is interesting given that today, we have more access to, and more ways to capture and store, project data than ever before. So does this mean that data has a detrimental impact on project performance? Does old-fashioned human expertise led by “gut feel” yield better results? Let’s take a look at the real story.
Building the Foundation
As noted above, one of the biggest trends to sweep industries across the board over the past few decades has been the use of technology to capture massive amounts of data right down to the most minute detail. The anticipation and excitement of mobile devices, GPS tracking, IoT, drones and machine learning have led to copious amounts of data — but the question is, what to do with it. Although the construction industry has been somewhat slower to embrace the data trend, there is no shortage of technology on every construction site.
Transforming this data into organizational knowledge can be useful in identifying trends, allocating resources more effectively and proactively preventing potential trouble spots before they happen. This data collection represents a critical first step on the road to better strategic decision-making on projects.
Unfortunately, gathering and storing large amounts of project data in and of itself means nothing. In fact, data can be quite overwhelming and counterproductive if not put to use appropriately. What’s needed is an understanding of how those data points relate to the bigger picture — insights that can only come through expert analysis. The key is to collect the critical data and connect it to other relevant data points throughout the project life cycle.
Data + Experience = Learning
This is where human expertise comes into play — understanding which data is relevant, which trends are meaningful and how to act on them to improve performance. Also, how to compile and express data in such a way that it’s easy to gain insights and gauge possible courses of action from there.
For example, committing to estimate a new opportunity can be an expensive endeavor. How do you know which opportunities to chase and which ones to let go? Certainly, client relationships and industry experience are important factors. However, other factors such as historical wins and margin on similar jobs are equally important. Creating a model that supplements human expertise with objective data is the key to chasing the right work and successfully building that work when the estimate is won.
All of these things require intelligent analysis, plus organizational knowledge of past projects and the data and trends associated with them, in order to act effectively. In this way, the power of data combined with human experience can be invaluable.
The Secret Ingredient: Unlearning
Now for the crucial and often overlooked third piece of the puzzle in the data/experience equation: The biggest pitfall in applying human expertise to the analysis of data is that bias will inevitably creep in. Those using data with the intent of improving project performance often look at the data they collect to confirm their opinions rather than using it as a basis to question their opinions and acting on findings from there.
This raises a key point — that you should analyze data not just to support your proposed plan of action, but also to try to prove it wrong. Using the estimate example from above, it is critical to factor in previous jobs, similar in size, location and market that didn’t go so well, or estimates that were pursued but not won.
This cross-section of data points provides important data along a positive/negative continuum that can be considered along with personal experience. It could be that it’s best not to pursue that job that sounds enticing. Perhaps in looking at the numbers you can see that, in the final analysis, the time and resources spent would be detrimental to your business and the bottom line.
Many of us don’t want to acknowledge the implications of the data we collect and analyze if it conflicts with our preconceived opinions and processes. But as leaders, we need to become fully comfortable with being wrong, and even embrace the opportunity to “unlearn” what we think we know based on data.
Only when we acknowledge both sides of the coin can we truly leverage the value of data, and the analysis that goes with it, to achieve our full potential on projects.