Telling the story with data.
Read enough analyst reports, and listen to enough conference speakers, and one can be forgiven for assuming that all important decisions these days should be data-driven. However, business leaders and professionals should be selective as to which decisions should be tied to data analytics, and which should remain based on human experience.
For the most part, low-level decisions — such as recommending upsells or extending credit — have have been great candidates for automation and AI-driven decision-making. The next great frontier is higher-level decisions that may be more strategic to the business. We are at a crossroads, and may be ready to turn over more strategic decisions to the machines — or at least get the machines to help. And this is where things get rather complicated. Can AI really make the call whether to staff up a struggling contact center? Or to outsource a production line?
To help make decisions about making decisions, Gartner recently published an informative ebook, The Future of Decisions, that outlines the process of dissecting and reengineering decisions to both separate human input from machine involvement, as well as enable the two to work in unison.
The book’s author, Gartner analyst Patrick Long, says there are three levels of engagement for humans versus machines in decisions:
- Higher-level decision-support: Decisions made by humans, may be “based on principles and ethics, experience and bias, logic and reasoning, emotion, skills and style,” Long writes. Machines can “provide visualizations, exploration, alerts and other support for human decision makers.”
- Augmented machine support: This may be based on augmentation where “machine suggests; human decides”; “human suggests; machine decides;” or “human and machine decide together,” says Long. The role of machines and AI in these decisions is to “generate recommendations, provide diagnostic analytics for human validation and exploration.”
- Highly automated settings: There is still a need for “guard rails or a human-in-the-loop for exceptional cases,” says Long. Otherwise, machines can handle autonomous decision making “using predictions, forecasts, simulations, rules, optimization or other AI.”
It’s time to re-engineer decision-making to make them more data-driven, Long urges. “No decision stands alone. Decisions by one actor affect other actors in the enterprise and ecosystem, and vice versa.” To get there, it’s important to understand what decisions are important to the organization. Take a deep look at how decisions are being made, and conduct interviews with executives and managers, or even sit in on meetings to document how teams arrive at decisions.
After studying colleagues’ decision-making processes, determine how data analytics or decision intelligence technology can fit in and make a difference. This can include “augmented, diagnostic or predictive analytics,” says Long. The data architecture and infrastructure also need to support increasingly data-driven initiatives — preferably, a data fabric approach that also includes metadata management, data catalogs, and business glossaries or ontologies.
It’s just as important to develop peoples’ analytical skills to adapt to an increasingly data-driven decision-making environment, Long says. Organizations may even want to create the role of “decision engineers” who will “diagnose and rethink decision-making processes.”
