The bottom line of that presentation was that the fault lies with corporate leaders who don't understand big data, statistics, and analysis. Using a series of quotes that I have had the pleasure to hear in person, I called out some of the errors and assumptions that executives make when being presented with a model or analysis. These errors undercut their attempts to make data driven decisions.
What I did not have time to do in that presentation was to peel back one more layer of cause to show that the executives are as much victims of a poorly constructed system as they are perpetrators of bad decisions.
Let me illustrate by relaying what actually happens in meeting with a chairman, presidents, and senior vice-presidents.
You have these key players together for about 2 hours. During which time each of them will receive somewhere around 200 emails updating them on various initiatives, projects, and ongoing concerns. Somewhere in the email will be a handful of requests asking them for permission to start some new project or spend some amount of unplanned money because of a change to the industry that requires an un-looked for response.
While they are sitting there with phones vibrating softly to let them know that the business is falling apart without their immediate attention, the other executives are asking for their partnership ("please share my costs") or alerting them that there have been changes to an ongoing program ("the sky is falling") or asking them for updates ("who should I blame for the drop in profits").
Into this mix of divided attention, blame-shifting, and contextual negotiation walks an analyst who needs to provide the results of a test and ask for a decision. The analyst has been told that these are important people with very little time and that they can only have a 3 page presentation with fewer than 3 bullets on each. And throw in a chart to make it colorful. The result is a vague, high-level description of something that is detailed and complex.
Some background on being an executive: no one ever comes to you with a bad idea, but some ideas are bad. After about 32 seconds of being a vice-president, you realize that some of the glowing recommendations that you are being provided have to be false. You can't tell which ones are false, because they are all 3 pages with 3 bullets per page and a pretty chart, which is not enough detail to do a reality check. So you develop the habit of asking what you hope are penetrating questions to get to the heart of the matter, or you make summary statements to see if you are understanding the material. When we talk about "good executives", typically we mean people who can scape off the icing to see whether there is cake or mud underneath with just a few questions. "Bad executives" scrape off the icing and find that there is icing underneath. "Geed news, the future is filled with delicious icing for ever!"
Back to our analyst. He's got his 3 slides and 3 bullets per slide. He has spent the last 6 months working on this project and is pretty pissed that it all comes down to five minutes in front of people who can't even be bothered to put their phones down.
And then someone asks him a question.
Joy! They care about his work. They want to know more. The question was kind of off-target, but it gives him a chance to tell them about the next layer of detail that they really need to understand. He answers. A few of them look up from their phones and seem non-plussed. Someone asks a follow up question that seems just like the first one to the analyst. He gives them more detail.
To the executives, someone walked in with a pretty presentation and a rosy recommendation. They asked a couple of questions ("Will this margin drop to the bottom line as profit? If the projections are off by 10%, how much does that impact the projected results?"), and the simple presentation has turned into a crazy Rube Goldberg machine.
It would be easy to say that the problem with Data Driven Decision Making is that we are not training our executives. In fact it was easy to say. I said it last Sunday, and I didn't break a sweat. But that is an oversimplification.
The next layer of cause is that we are relying on a handful of people to be experts in everything. If it makes you feel any better, we have only been struggling with this problem since, you know, forever. Ask Sargon how he made decisions about palace construction and you will see what I mean.
How do we fix it? What's the solution? The only thing that I can see is that we abandon "leaders" in favor of "facilitators". Move decision making down to the level of expertise and charge executives with bringing the right experts together and ensuring that projects have appropriate support, don't charge with making the actual decision. Will that happen? No. It would take a core change to human nature. We would have to be willing to say that the president of the company can't be held responsible for the company's decisions. We aren't ready to say that Obama isn't really in charge and that a couple of hundred health care analysts are actually responsible for the Affordable Care Act.
There's your happy thought for the day!
No comments:
Post a Comment