5 Ways that Wind is Getting Big Data Wrong

When it comes to capturing value from data, the windpower industry remains far behind its industrial peers. Owners and manufacturers don’t suffer from a lack of wind data, and many are using software and physics models to monitor and generate predictions with potential to impact the bottom line. Yet with a few exceptions, the sector has yet to fully capitalize on profit gains from data analytics.

Here are 5 ways that wind is getting big data wrong:

1. Focusing on data instead of business value. This is the typical approach we see in the wind industry: first, collect as much data as possible. Then, hire a bunch of PhDs to generate some algorithms and share the resulting insights via business intelligence tools. And finally, hope that your employees will make smarter decisions.

If owners were learning from other data-intensive industries, they would begin the process by prioritizing specific business outcomes and then they would work to understand which people and workflows have the greatest power to create those outcomes.

Next, they would develop data science approaches and software with specific users and workflows in mind. And finally, they would close the loop by tracking impact and use machine learning to improve algorithms based on the results.

2. Oversimplifying the meaning of “value.” When wind companies attempt to create value from data analytics, they often fail to separate and prioritize the potential value opportunities. As a result, they spend millions of dollars creating mediocre internal tools and commercial products.

For example, time-based availability may be a worthwhile goal for an owner. But how does it relate to increased performance and energy capture? And what about availability when market prices are high?

Until you understand which specific problems have the biggest potential to drive value, you don’t know which data analytics solutions can unlock that value.

3. Conflating and confusing different solutions. When most wind leaders use the term “analytics,” they are referring very broadly to data collection and subsequent reporting functionality.

Yet to a data scientist, there are massive differences between simple business intelligence dashboards and things like anomaly detection algorithms that diagnose problems and prescriptive models that recommend action. Each one requires a specific set of data and often a completely different technological approach. It’s crucial to understand the interaction between available data, high-value problems, technological feasibility, and potential solutions.

4. Neglecting to tie insight to action. Most wind software fails to tie data-driven insights into workflow so executives, analysts, managers, and technicians can quickly make better decisions and take effective action. That’s because wind software designers have largely failed to conduct user experience (UX) research to map the details and pain points of existing workflows.

User-centered software design, which draws on this kind of ethnographic research, has been shown in many industries to enable frictionless user adoption and more effective engagement during design and deployment. For wind, it’s the best way to harness the full value of data and expertise to improve performance.

5. Trying to be software companies (which they’re not). Leaders in the wind industry are smart–they see that to survive and thrive, their companies must evolve into digital versions of their former selves.

But data and software are not their areas of expertise. So they end up spending millions of dollars hiring people to create internal tools that don’t work and don’t scale. Or, they lay out cash to a data strategy consultant who tells them which system integrator to hire, which point solutions to buy, and which change management consultant to hire to change their employees’ business processes around the software (instead of the other way around). Even when they create great analytics insights, they don’t have a platform that can deploy them into workflow quickly enough to drive value.

The alternative to these traditional methods is to employ a more collaborative model that allows industrial firms to capitalize on their core assets (subject matter expertise, access to data, and industry leadership), while benefiting from the data analytics platform and technology skills of an agile, entrepreneurial partner.

At Uptake, we have pioneered this method and we call it “Collaborative Disruption.” While we’re not the only company that offers these capabilities, I believe we’re the most effective. That’s why I left an executive role at one of the world’s largest energy companies to lead the Uptake Energy team.

I invite you to contact us to talk more about how our predictive platform is changing the wind industry at uptake.com/wind.

Sonny Garg is a seasoned executive who has led organizations through critical points of inflection, including large-scale integrations and disruptive innovation.

Prior to Uptake, Sonny spent 13 years at Exelon Corporation and served as the President of Exelon Power.