The insight-driven organization

Short Takes...on Analytics

A blog by Tom Davenport, independent senior advisor to Deloitte Analytics​.

Get serious about insights

​​​It’s not unusual these days for organizations to want to make decisions on the basis of rigorous, data-based insights rather than experience or intuition. In fact, my friends in the Deloitte Analytics practice have begun to talk about the “insight-driven organization,” or IDO. I am not closely involved with that set of ideas, but I thought I would use it as a provocation to dwell on what it can mean to get serious about insights.

“Insights” can be defined in a variety of ways, but in this case I am referring to getting an understanding of the true nature of something by virtue of extensive and systematic analysis of relevant data. That “something” is most frequently a customer when we are talking about business insights, but it could also refer to a supplier, a financial event, or an employee. Having true insights on these entities means you are likely to predict their behavior—at least to some degree—and achieve a positive outcome.

We usually think the tough part of insights is creating them. But there is plenty said and written about that topic. We sometimes neglect the earlier and later stages of the process of “insight management.” Perhaps the overarching challenge is that very few organizations think about insights as a process; they have been idiosyncratic and personal. But if companies are serious about becoming IDOs, they should focus on how insights are framed, created, marketed, consumed, and stored for reuse.

Let’s assume that you know something about insight creation with data and analytics. But how is the insight framed in the first place? Good framing helps set up the entire process of generating and using insights. What problem needs to be solved? Is the decision being addressed amenable to being answered by data and analytics? Are enough alternatives being examined? Are the decision-maker and the analyst aligned on how the problem is framed and pursued?

Consumers of insights are most commonly decision-makers within an organization, and they should be carefully engaged with the process from the beginning. Martha Feldman and James March, in a 1981 article observed that the insight management process doesn’t always flow smoothly from available information.1 In fact, they observed that potential consumers of insights typically requested a lot of information that they didn’t actually use in making decisions. Because (even in 1981) analytical decisions were considered socially desirable, many managers tried to appear to be analytical decision-makers when they were really using their gut. This finding, which I feel is still relevant today, means that analysts should be sure of how the decision-maker plans to use the information before gathering and analyzing it.

Appealing to the consumer of the insights means, in some cases, using analytical approaches that are less than fully rigorous. Certainly some insights are sounder than others in terms of their analytics, but that doesn’t mean they will result in a better decision. I once interviewed a wise market researcher about the techniques he used with his clients. He was well aware that focus groups, for example, are not typically known for generating high-quality insights about customers. Focus groups are notorious for telling marketers what they want to hear, among other issues. But the market researcher sometimes employed them anyway, he said, if that was the only kind of insight that his client was prepared to act upon. He believed that insights based on questionable data might be better than those based on no data at all.

Many analysts feel that when the insights have been generated, their work is over. But there is a critical marketing step in the insights process. I first became oriented to this topic when reading the 2012 book by Marco Vriens called The Insights Advantage: Knowing How to Win, which offers an interesting perspective on the need to market and sell your insights if they are to make an impact.

Good marketing of insights often goes back to the framing process. Vriens notes, “Understanding what people need to accomplish is really the foundation of impact.”2 He mentions that impact may be achieved by affecting a particular decision, or by affecting the firm as a whole through improvements in financial or operational processes. Insights may also simply improve the organization’s understanding of what drives its performance in the marketplace.

Vriens notes that getting insights understood and accepted is another critical aspect of the insights process—as it is with marketing any other product, service, or idea. The decision-maker should know and understand what the insights are, that they have to be credible, and that the actions to be taken have to be clear. The insights should be contained in a well-told story if a human is going to internalize and adopt them.

Finally, there is the issue of what happens to a particular insight once it has been created, digested, and applied in a decision. Is that the end of its life? Can subsequent analyses and decisions benefit from an awareness of previous insights?

It’s likely that they can, but few organizations seem to be aware of that possibility. An insight is a piece of knowledge about how your business works. If you want to understand how your business works in general, it’s often useful to peruse a variety of previously-generated insights, particularly the most recent ones.

I once worked with a bank that generated 60,000 insights a year, primarily through controlled testing programs. They did have a knowledge management database in which the insights were supposed to be stored. But executives there admitted that many insights didn’t make it into the database, and many people never perused past insights before designing their own tests. “People need to discover things for themselves,” one manager told me.

If that’s the case, it seems a little discouraging. It means that we may have to devote a lot more effort to insight creation and marketing than we would otherwise. If an organization cares enough about insights to treat them as long-term assets, it might sooner throw away money than throw away a useful insight. In short, becoming an insight-driven organization should mean not only being able to create and apply a lot of insights, but also being willing to hang on to and reuse them over time.

1. Martha Feldman and James March, “Information in organizations as signal and symbol,” Administrative Science Quarterly 26:2, June 1981, 171-186,

2. Marco Vriens, The Insights Advantage: Knowing How to Win, iUniverse, 2012, 80.

The insight-driven organization: Management of insights is the key" was published in DU Press on August 26, 2015.​​

Back to top


"Very few organizations think about insights as a process; they have been idiosyncratic and personal. But if companies are serious about becoming IDOs, they should focus on how insights are framed, created, marketed, consumed, and stored for reuse."

More about the author, Tom Davenport.

Did you find this useful?