W. Edwards Deming famously remarked that “in God we trust, but everyone else should bring data.”  It has come to typify a booming industry in predictive analytics, with estimates suggesting the sector will be worth around $270 billion in 2022.

Despite this huge outlay, research from Rotman highlights the profoundly mixed returns seen on that investment.  The researchers found that the best-performing companies tended to make significant and complementary investments in their workforce, their IT systems, and their manufacturing processes.  When they did this, they were able to secure revenue increases of up to $1 million on average.

“These complements provide the organizational infrastructure to collect, analyze, and respond to predictions based on objective data,” the researchers say.  “IT capital captures investments in data collection and computer hardware that can transmit, store, and analyze data, for example. Educated workers are known to be an essential ingredient for that system. And certain production environments provide richer data due to the processes they use.”

Strong returns

The researchers assessed a representative sample of manufacturing plants in the United States in 2010 and 2015, with a survey asking them about their use of predictive analytics, their management practices, the use of data in their decision-making, and the general design of their production process.  The results were then cross-referenced with data, including production inputs and outputs for the firm.  The firms were chosen in part due to their propensity to be early adopters of technology.

By 2010, around 75% of the sample had adopted some form of predictive analytics, although most of these firms only used the tools sporadically, despite higher intensity use being linked with greater returns.

Despite repeated research urging companies to invest in skills development alongside IT investments, there appeared to be many firms that were still lacking in this regard, and were therefore not getting the returns on their investment that they might.

“We found it puzzling,” the authors conclude. “More research is needed to understand the organizational or market frictions that are causing this apparent misalignment, one that is proving to be quite costly in the firms we observe.”