We’ve written a great deal about digital civil rights. The typical citizen believes they have a right to transparency. That means they believe they have rights about how their data is shared, used and, how analytics affects them personally. We call this the “right to transparency.”

But what about business data? What do business leaders think they have a right to expect?

Obviously, the business questions are different than personal data. There are probably no personal privacy rights involved in the analytics of a firm’s warranty claims. But that data might benefit a competitor, so most firms will protect it.

What’s less obvious is the increasing demand for transparency, even when the data and analytics are shielded from public view. What does this executive right entail?

Our research shows the answers are much like the expectations of individual citizens. We conducted a survey of more than 200 executives. We asked them questions about their willingness to act on analytics results. However, in each case, we asked two versions of the same question.

Here is one example of a question pair;

Imagine you are leading a large organization providing services to support the hardware your company sells. You want to keep your customers happy and their hardware running. You also want to make sure you are profitable while you do that. Your business analysis team shows you an assessment. It claims you are buying far too many spares for some of your most expensive parts. It also suggests you have a few expensive items which should be purchased in larger numbers.

If you take their recommendation, and if they are right, you will have both a profit improvement and an improvement in customer satisfaction.

When you ask questions to determine if you think they are right, they tell you they have a detailed set of business rules and valuation methods. Their analysis is transparent and they can dive as deeply as you have time for.

We asked that same question, but substituted this, for the final paragraph:

When you ask questions to determine if you think they are right, they tell you they used a Convolutional Neural Network. It’s a “black box” and no one really knows how it works.

Willingness to accept the team’s recommendations more than doubled with transparent analytics.

Perhaps as important, the respondents were four times more likely to completely reject all the recommendations in the “black box” instance. This is bad news for the neural network, because it’s very difficult to achieve even partial transparency for the CNN.

Using a different scenario, we asked respondents to imagine they were making a critical strategic corporate move when a competitor had filed bankruptcy. In this scenario, moving fast was critical.

Our executive respondents were five times more likely to reject the team’s recommendations for the black box case, compared to the transparent analytics team.

So, what is the executive’s right to transparency? It seems to be about what GDPR and the new, similar law in California say about individual rights.

  • A right to understand what algorithms are doing
  • A right to understand what “the answer” is
  • A right to connect the dots from data to model output

Even in marketing, the problem of transparency persists. In June, a survey of business-to-business (B2B) marketers from North America showed that only about one-in-five felt they understood AI, Machine Learning, and Predictive Modeling. Only about one-in-eight were confident about their knowledge of AI in marketing technology. So, not surprisingly, a large majority of marketing professionals don’t use AI.

The marketing survey is important. Everyone knows a great deal of marketing targets are flawed, sales calls are dead ends, and advertising misses the mark. Marketing involves high failure rates. And, no one can write a physics equation for buyer behavior. Surely if there was a sweet spot for AI, this would be it.

The right to transparency doesn’t matter much for facial recognition, product recommendations, and natural speech interfaces. Neural Nets are great at all those things, and we have a hard time thinking of something better than being data driven in those cases.

But this new research suggests a law of human behavior for analytics: The Glass Box Rule – More important decision demand more transparent analytics.

The Glass Box Rule helps explain these results: executives get paid to make important decisions, it’s no surprise they expect transparency. https://www.lone-star.com/insights/blog/

About Lone Star Analysis

Lone Star Analysis enables customers to make insightful decisions faster than their competitors.  We are a predictive guide bridging the gap between data and action.  Prescient insights support confident decisions for customers in Oil & Gas, Transportation & Logistics, Industrial Products & Services, Aerospace & Defense, and the Public Sector.

Lone Star delivers fast time to value supporting customers planning and on-going management needs.  Utilizing our TruNavigator® software platform, Lone Star brings proven modeling tools and analysis that improve customers top line, by winning more business, and improve the bottom line, by quickly enabling operational efficiency, cost reduction, and performance improvement. Our trusted AnalyticsOSSM software solutions support our customers real-time predictive analytics needs when continuous operational performance optimization, cost minimization, safety improvement, and risk reduction are important.

Headquartered in Dallas, Texas, Lone Star is found on the web at http://www.Lone-Star.com.