It looks like nothing was found at this location. Maybe try a search or browse one of our posts below.

Whirl a ball on a string (in outer space, where gravity has negligible influence). Let go of the string. How does the ball fly?

Ask a child, or an adult who has little formal education in physics, and they’ll tell you the ball will follow a curved path. Anyone who knows Newton’s laws of motion, or who has whirled a ball on a string and let it fly, knows that isn’t the case. The ball’s trajectory will, of course, be a straight line.

If trusting your gut gives you the wrong answer for a question this simple, it’s unlikely to be more reliable when you’re faced with a more difficult challenge — say, selecting a piece of software for the enterprise. Yet recent research by psychologist Ap Dijksterhuis of the University of Amsterdam and a team of colleagues provides evidence that the opposite is the case.

Dijksterhuis created two car-buying situations — in each, college students had to choose from among four fictional automobiles. In one, they were given four attributes. In the other, they were given twelve.

In each case, Dijksterhuis distracted half of the research subjects during the four minutes made available for the decision (Dijksterhuis apparently likes the number 4), while allowing the others to spend their time thinking. With four attributes, thinkers slightly outperformed non-thinkers. With twelve attributes, those prevented from thinking made, on the whole, better choices.

It would seem that, faced with a choice between SAP and Oracle, you’d do best by having each company throw a lot of facts at you while you’re thoroughly distracted by other matters — perhaps, a crossword puzzle. As soon as you finish the puzzle, you should choose whichever piece of software seems best to you.

I don’t think so. But if this isn’t the case, how should we interpret Dijksterhuis’s research?

A definitive answer will require more research. In its absence, here’s how I’d assess the situation:

People aren’t born knowing how to make evidence-based decisions when multiple factors are involved, and few students receive formal training in it. Why would we? It’s merely one of the most important life and business skills we all need on a regular basis.

This explains the superiority of “instinct” (really, non-linear, pattern-based decision-making) in Dijksterhuis’s experiment.

Most people, faced with a decision, don’t put much thought into how they’re going to make it. Even if the 80 college students who made up Dijksterhuis’s sample universe were the exceptional sort who do, the experiment created an artificial situation that prevented it from happening.

There’s a difference between having time to worry about a decision and having a planned framework for making it. Look at how the research was constructed. The investigators pushed a lot of information at subjects who were unprepared to receive it.

Compare that circumstance to how you, as an IT professional, would go about making an important decision. First, you’d create a framework for making it. You’d establish categories of information, specific questions for each category, and figure out in advance which answers to each question are preferable. You’d take control of information-gathering so you’d be prepared to record and understand the answers. Probably, you’d establish a way to turn answers into numbers and aggregate the results, to facilitate comparison of the alternatives.

Then you’d make a rational, informed, evidence-based decision.

In Dijksterhuis’s experiment, it isn’t all that surprising that, faced with information overload and a lack of clarity about what’s important and why, time simply created more opportunity for second-guessing. As is so often the case, the results of scientific research (information overload creates confusion) were used to reach a conclusion that’s more appealing than warranted (gut instinct is superior to thinking).

Is this the fault of the researcher? In this case, yes. Those who perform research are best equipped to explain its limitations, and are responsible for doing so. It appears Ap Dijksterhuis has done the opposite: He sensationalized the results, extending the mythology of human “instinct” (a mis-use of a term better reserved for inborn knowledge) in the bargain.

You can be sure that sometime in the not-too-distant future, your life will become more difficult in consequence. The popular and business press have reported this research, so it’s only a matter of time before a swarm of business consultants, eager to pander to those inclined to buy advice that fits their preconceived notions, will sell “instinctive decision-making for business executives” to their clients.

And intellectual relativism will have made another inroad into a society already predisposed to favor personal preference over evidence.

Depending on the business expert I’m listening to and the day of the week, I know three truths:

1. Good employees who work together as a team outperform great employees who don’t.

2. Good employees with great processes outperform great employees with bad processes.

3. If an employee is irreplaceable you should immediately fire that employee.

From first-hand observation I know that when it comes to Information Technology organizations:

  • Great employees can and do overcome bad processes.
  • Great employees can and do overcome lousy managers.
  • Great employees can and do pull along mediocre teams.
  • Making one or two great hires is the most critical step in turning around an underperforming organization.
  • Well-designed processes can be pretty useful, too.

Which is to say, if you want an organization that works, you’ll get more leverage from hiring great employees than from any other single effort you can undertake.

Great employees can overcome organizational deficiencies to deliver useful results. They can’t, by themselves deliver a great organization. That takes a lot more.

The question of what makes a great organization tick is rife with superficial thinking. The usual approach is what you might call the “Tom Peters Fallacy”:

  1. Find a great organization.
  2. Identify a trait in that organization you like.
  3. Decide that this trait is what makes that organization great.
  4. Declare that this trait is the panacea for all other organizations.

As far as I can determine, there is no one characteristic that by itself can make an organization great.

Well, okay — there is one: excellent leadership. That only works because I define “excellent leadership” as “Doing everything required to build a great organization,” thereby begging the question.

So … what is required to make an organization great, as opposed to simply functioning?

Leadership: A great organization does start with strong leadership, in a non-question-begging way. If there’s no direction — no focus, no goals, no plan, no definition of excellence, no clearly stated expectations for employees to live up to; no alignment of purpose and standards — the employees will keep things going, but not much more than that.

Great employees: Not every employee has to be a superstar, although all must be competent. Great organizations do need enough top-notch performers to demonstrate that high standards are achievable, not theoretical.

Focus on achievement: The definition of “great employee” has been diluted through too many managers reading about “emotional intelligence.” Employees who are focused on getting along will concentrate on how irritating their colleagues are. Employees who are focused on achievement will value their colleagues’ contributions and ignore their eccentricities.

Teamwork: Just as the definition of “great employee” can’t ignore the importance of serious technical ability, it also can’t ignore the importance of working and playing well with others, and of providing leadership in the trenches.

Willingness to innovate: This is IT we’re talking about. Information technology. The field where if you can buy it, it is obsolete by definition. IT organizations and everyone who works in them must be willing to try new technologies, processes and practices … and even more important must be driven to constantly find improvements to the ones already in production … or they stagnate.

Willingness to not innovate: “State of the art” means “doesn’t work right yet.” Most of the time, the work required of IT is best achieved by extending what you have, not by chasing whatever is being hyped in this year’s press releases, aided and abetted by publications hungry for advertising revenue.

Evidence-based decision-making: Great organizations make decisions through the use of evidence and logic, not wishful thinking and listening to one’s intestines.

You’ll note the usual buzzwords are notably absent. Governance, ITIL, CMM and all the other processes and practices (I used to call them Processes and processes) that are supposed to lead to inexorable success don’t, in fact, lead to excellence.

Excellent IT organizations do have them. Even more important, they are kept in their proper place — as useful tools that help employees be as effective as possible.

The best woodworkers have band saws, coping saws, lathes and routers in their workshops, and not just hammers and chisels, but their tools aren’t what make them the best.

Great IT organizations are the same. They practice good governance; follow consistent application maintenance, enhancement, design and testing methodologies; adhere to clearly defined change control procedures; and otherwise avoid making things up as they go along.

Their processes and practices are important. They are, however, merely the signs of a great IT organization.

They are not its cause.

ManagementSpeak: Now that you’ve secured our network, we’d like to start examining our vulnerabilities in some non-traditional areas.
Translation: I need a scapegoat for all sorts of problems, and since I don’t understand this technical stuff anyway, you make a good target.

Speaking of vulnerabilities, this week’s contributor figured his would be reduced by remaining anonymous.