It’s safe to say the application of data analytics to test & measurement (T&M) is lagging behind what’s happening in the financial, insurance and retail industries. However, as an optimist, I view this situation as a greenfield opportunity for you and for us. The ultimate benefits are making better decisions in less time, delivering the right product, and getting to market faster.
This is also a good time to pause and define what you and your organization aspire to in the use of data analytics (DA). Your goal may be as bold and transformational as overhauling your product lifecycle (albeit gradually). Or it may be as basic and personal as achieving tighter control over your projects—and, in all honesty, this is what I aspire to.
Addressing the fear
Another confession: While I have a deep appreciation for the benefits of DA, I dislike the term. The reason: “data analytics” can feel so overwhelming that it produces pangs of fear among many of my colleagues and some of our customers.
To narrow the scope, we can reframe it as “data analytics for test and measurement” (DA for T&M). And we can think small: you can use DA for T&M as a way to validate your design intent. Here, the goal is to gather the right data in the right amount to say, with confidence, “Yes, this product delivers on our design goals and we can build it reliably.”
Determining your team’s maturity level
Before diving in, it’s useful to step back and assess your team’s maturity level with DA. In this context, I see four levels: reflect, predict, prepare, and influence.
- Reflect is a look backward at what has happened in the past, aiming to describe and understand the root causes of recurring problems and, if you’ve been collecting data, investigate any troublesome trends.
- Predict builds on reflect by creating a working theory and then outlining scenarios for what is likely to happen internally and externally.
- Prepare is the first step toward an active attitude: given a prediction of what is likely to happen, the organization takes steps to be ready for those situations.
- Influence is a deeper engagement with the active attitude: if predict and prepare are pointing toward unfavorable scenarios, the organization takes steps to nudge the variables toward a more beneficial outcome. This is when we start aspiring to greater levels of control.
Taking your first steps
From my conversations with customers, most are like us: somewhere between “Where do we start?” and “How do we move beyond reflect?”
It isn’t about moving mountains, at least not at first. Start with the end in mind by defining the question you’re trying to answer. Then, carefully design the “experiments” you want to conduct and make a plan based on the type and amount of data you need to measure.
Next, be sure you have the tools to make the necessary measurements (actual values). Ideally, you will also be able to capture design simulations (predicted values) prior to building anything. Even if you can address only the validation phase—between design and manufacturing—this is still a worthwhile activity.
As a practical tactic, apply some up-front organization: Where should everyone store their data? What format should they use? Who will manage it? What common tool will they use for analysis?
Then get started by making measurements and collecting data from a batch of prototypes or pilot-run units, tested across a sufficiently broad range of properties and operating conditions (yes, this could yield several thousand data points). As you (and your team) are collecting data, sift through the initial raw data, apply “sanity checks” and, if everything looks good, start analyzing. This works best if your team has a common, shared tool that gives everyone access, enabling them to view data through their individual “lenses” and flag anything that seems out of kilter. By enabling your team to monitor progress together, you can adapt to insights and optimize your efforts and your results.
For project managers, I suggest using the early results to make a few less-than-critical decisions. Then, take stock: How much is it helping? Are we spending less time while making better decisions? What can we tweak to improve our DA process and results?
That’s how you start climbing the maturity curve.
Pushing through the fear
For me, brief moments of anxiety still occur when my manager asks, “What’s the basis of your decision?” No one has complete control of their project—but, thanks to DA, my angst is subsiding because I can confidently respond with a credible answer.
When you push through the fear, there are meaningful rewards on the other side. These include making better decisions in less time, delivering the right product, getting to market faster... and sleeping better (most of the time).
My next post will drill down to the next layer of the story, offering tips that will help you extract insights from your “data mess.” In the meantime, let’s discuss: Have you already started down this path? If so, what is or isn’t working? If you haven’t, what’s holding you back?