Before McKinsey & Company senior partners Scott Keller and Mary Meaney decided what to include in their new book about organizational leadership, they grouped all of the leadership-related articles published by Harvard Business Review from 1976-2016 into 20 topics--culture, change, self-improvement, managing others, etc. Then, they analyzed how the number of articles written on those topics varied over time as a percent of all HBR articles.

"Our logic was that the lower the variance over forty years, the more timeless the topic," write Keller and Meaney in their introduction to Leading Organizations: Ten Timeless Truths (Bloomsbury, June 2017). The most timeless topic of all? Decision-making.

This didn't particularly surprise the consultants. They calculate that there are roughly 400 million decisions made every day in the average Fortune 500 company. Of course, the vast majority of those decisions are inconsequential, but a select few of them could be very consequential indeed.

Like, say, Time Warner CEO Gerald Levin's decision to merge with AOL back in 2000--on the eve of the dotcom meltdown. In 2003, the merged company posted a $99 billion loss on what is still widely hailed as the worst deal ever. The flamboyant Ted Turner, who was Time Warner's largest shareholder before the deal and lost about $8 billion personally, likened it to the Vietnam War on his personal disaster scale.

To avoid disasters like this and hone your decision-making process and prowess, Keller and Meaney offer the following advice:

Decide what kind of decision you've got

The McKinsey partners say that there are three decision types and that each of them require a different approach.

Type A decisions are infrequent, high-stakes decisions that affect the enterprise broadly. "These decisions warrant disciplined decision-making processes that ensure the right data is gathered, open dialogue and debate take place, and all stakeholders' voices are involved and aligned to the greatest extent possible," write the authors.

Type B decisions include repetitive decisions that affect only one area. "These decisions should be delegated to the lowest possible level with clear and transparent accountabilities," say the authors.

Type C decisions are repetitive, cross-cutting decisions. "These decisions are not, in fact, one decision, but instead the culmination of a chain of smaller decisions which must be coordinated and aligned across organizational boundaries," explain the authors. "As such, a formal process that clearly defines decision protocols, particularly at key interfaces, should be put in place across the end-to-end system. These processes should also have built-in feedback loops to enable the process to be systematically improved over time."

Create a dialogue

Data is often named as the most important component of decision making. But when Keller and Meaney analyzed a study of 1,048 major decisions made over five years, they found that dialogue is six times more important than data in reaching sound decisions.

"While the above study was of type A decisions, our experience is dialogue is important across all decision-types that matter in an organization, though it looks different for each," they write. "For type B decisions that are delegated to individuals or natural working teams, this is why it is vital not just to assign the accountability, but also to be clear on who needs to be consulted by that individual or group in advance of making the decision. When it comes to type C decisions, dialogue is vital as it ensures that the needed points of alignment across organizational boundaries happen efficiently and effectively."

The consultants suggest using the RACI model to ensure that dialogue is driven into your decision-making process. It clarifies who is responsible for making recommendations, who is accountable for making the decision, who needs to be consulted before a decision is made, and who needs to be informed after the decision is made.

Beware of cognitive biases

We all fall prey to what Kurt Vonnegut called "our big brains." Wikipedia lists well over 100 kinds of cognitive biases.

Keller and Meaney call out three of the most common: confirmation bias, in which we give undue weight to information that confirms what we already believe; social bias, in which we give greater weight to what we think others will view favorably; and optimism bias, in which we expect the best possible outcome against all odds.

A diverse team is your best bet for avoiding cognitive traps--such teams improve decision making by as much as 50 percent, according to the authors. In addition, they write, "there are a number of proven and practical tools to minimize biases in decision-making including, among others, the following: the 'pre-mortem' (generating a list of potential causes for failure of a recommendation and working backwards to rectify them before they happen); 'red team-blue team' (assigning one person/group to argue for, and one to argue against, a decision); 'clean sheet redesign' (developing a system from only a set of requirements, free from considerations related to current investments or path); and 'vanishing options' (taking the preferred option off the table and asking, 'what would we do now?')."