How do we evaluate the decisions we make? Considering that decisions are essentially choices we make under risk and uncertainty with incomplete information, how do we tend to view our decisions subsequently with the passage of time, when the information we have is fuller and cold facts eliminate that uncertainty? When is a decision a good decision?

There are broadly two schools of thought on this. For many people, the instinctive response is that good decisions are those that lead to good results. If I chose to switch jobs, and the second job turned out to be great, it was a good decision; on the other hand, if I was worse off after switching, then my decision wasn't a good one. Indeed, the way we refer to decisions in casual conversations almost exclusively focuses on the results (Smart move, Wise choice and so on..)

On the other hand, others (usually on reflection) come up with this response – good decisions are those where the decision maker followed an appropriate process for arriving at the decision. That the manner in which the decision was made– the right data / information, the right considerations and appropriate logic – was correct. It was the right call to have taken at that point, even if things didn't work out later.

Let's say I need to get to a meeting at the other end of the town. I know that there are two ways to get there – one by using the main arterial road, which is slightly longer but where traffic moves faster; the other by smaller roads through suburbs, which is shorter but likely to be more congested. My history of prior experiences tells me that the arterial route is likely to be 10% faster. I consider all relevant factors (time of day, any recent construction activity or other such impediments I am aware of) and decide to take the arterial route. Turns out that there was a major hold-up due to some unforeseen reason, and I reach later than I would have if I had chosen the other route.

How good was my decision? Is it good because I followed the right process or bad because the results were bad?

(I know what my normal reaction would be – I'd be kicking and cursing myself, especially if I were running late for the meeting. It takes an unusually stoic temperament to remain detached and objective in this situation)

Ideally of course, we'd like to have both. If we follow a thorough process and arrive at a decision that also turns out well, nothing like it. The problem is we often get one or the other. Sometimes we get great results, with not-so-great processes due to a fortuitous turn of events. Other times, we do all the right things to no avail.

Our innate tendency is to evaluate based on outcomes. How often do we see this in say, cricket matches. The fielding side gets a couple of early wickets, the middle order batsmen dig in and are threatening to forge a partnership that could be match defining. The fielding captain brings in a part time bowler (to give his main bowlers a bit of a break, or just to break things up, or whatever..) for a couple of overs. Or perhaps he changes the field settings. And what do you know, the part-timer gets a wicket, or the batsman offers a catch straight to the fielder in the new position. “Brilliant captaincy!” say the commentators, and we agree. Of course, if the part-timer were to get clobbered, or if the catch were to go to the old position, we'd equally rail at the deficiencies of the chap wearing the captain's arm-band. I'd like to see one of the expert commentators evaluate the decision at the time it was made, rather than post-facto. In fact, some columnists make match predictions in various sports, and their hit-rates are no better than yours and mine would be. This doesn't stop them from carrying critical pieces simultaneously, which contain several post-match assessments of the 'woulda, shoulda…coulda…' variety. But then this is perhaps only to be expected – true foresight is as rare as hindsight is universal.

On the balance, my view is that the process is more important than outcomes. For one thing, we can only fully control our actions – we can't fully control outcomes. There are always going to be variables that come into play (call it Fate if you will) which we have no control over. Secondly, only a good process guarantees repeatability of overall consistently good results, as opposed to a one-off success. A blindfolded chimpanzee throwing darts at stocks could perhaps outperform an average fund manager using all the available information, models and decision making tools; but I'd lay odds on the fund manager doing better in the long run. (This example of course, takes me into highly unfamiliar territory – so I'll caveat it by saying that this assumes that both the stock market and fund managers have some method in their apparent madness)

This also places in focus the importance of building in learning mechanisms into a process. A process that remains static and doesn't learn from its history of wins / losses, is little better than trusting to blind luck. The next time I decide on the road route, I need to factor in the latest experience, attach a suitable weightage, not get over-influenced by the recency of the good / bad result and so on.

But if process is all that matters, doesn't this lead to the risk of people abdicating accountability, claiming that all outcomes are determined by Fate? We do know result-oriented people, who 'do whatever it takes' and stretch themselves to achieve seemingly impossible tasks, when it appears that the odds are against them. We also know people who operate purely in 'execution mode', doing what the Standard Operating Procedure demands and nothing more, and 'explaining away' outcomes. Will everyone not end up like the latter, following this school of thought?

This is a tough one, to be honest. As a practicing manager I tend to follow the 'Measure Outcomes, rather than Activities' approach. My reasoning is that measuring outcomes allowed personnel to grow - figure out solutions themselves, rather than be spoon-fed; take ownership, responsibility and accountability for their work areas. Isn't this at odds with my tilt towards process over results? Can a commitment to measuring outcomes as a performance yardstick ever be compatible with a belief that outcomes are never fully controllable and that the process matters more?

I think both can co-exist (though you could get schizophrenic at times!). When affixing responsibility and setting down performance yardsticks, you must emphasise outcomes and accountability for results to drive the stretch. However, you must make it clear that these are not to be achieved at the expense of the 'right means'. In a sense, you are self-imposing additional hard constraints, and making the problem more difficult that it would be if you relaxed one or the other. Basically, no short-cuts. It is surprising how often human ingenuity manages to work things out even with the hard process constraints, once people recognize that the boundary lines are inviolable. Finally, when evaluating performance, you have to factor in the specific circumstances – including the variables that actually were in control, events that could not have been foreseen and the efforts that were actually put in, before concluding on whether or not the right decisions were taken and whether or not performance was actually good. This requires maturity, balance and detachment.

"May God grant me the power to change the things I can, the ability to accept the things I can't and the wisdom to know the difference."