The Efficient Light Bulb

Once upon a time, there was a light bulb. This light bulb was quite a remarkable light bulb: it was praised far and wide for its incredible efficiency. This light bulb gave off no waste heat. This light bulb did not contribute to global warming. It had no carbon footprint. It did not rely on fossil fuels. Truly, it was an amazing light bulb and visitors came every day to see this remarkable light bulb.

One day, though, a traveler coming to see the light bulb in action was delayed by an unfortunate flood that closed several roads. He did not arrive until well after night had fallen. Much to his surprise, he found the light bulb sitting in a pitch dark room.

“Why aren’t you giving light?” asked the traveler.

“Give light!” replied the light bulb in shocked tones. “You must be joking. If I did that, I would use fossil fuels. I would have a carbon footprint. I would give off waste heat. I would no longer be efficient.”

“But isn’t the purpose of a light bulb to give light?” asked the traveler.

“I’ve always been told to be efficient,” replied the light bulb with a shrug. If you have never seen a light bulb shrug, it is truly a wonder to behold. The traveler would have been amazed, except, of course, that the room was too dark for him to see the miraculous event.

Once upon a time, there was a software company named “Soak, Inc.” Soak’s product relied upon a very complex database server. One day, the VP of Engineering stormed into the office and declared, “The server is too slow. We need to speed it up.”

From that day forth, every effort was focused on improving the speed of the server. Other issues were deemed insignificant beside the one, critical, goal of performance. Engineers who dared to raise other issues were publically humiliated for wasting the company’s time. Bugs that did not relate to performance issues were deemed “optional.” People who spent time reviewing the optional bugs and trying to fix them were warned that their insubordination would cost them their jobs if it did not cease immediately.

Eventually, Soak developed an amazingly efficient server. It was fast. It was robust. It was ready to demonstrate to potential clients.

The demo started out remarkably well. The server did not crash, causing some to believe that this couldn’t actually be a demonstration of a software product. Indeed, the server performed flawlessly. All would have gone well indeed for Soak had not someone noticed that the data being delivered by the server didn’t make sense. Yes, what the server had gained in performance it had lost in accuracy. In other words, it was incredibly good at very rapidly delivering useless or incorrect information.

When the engineers were questioned about this unfortunate oversight, they shrugged and replied, “We were told to be efficient.”

While it is not nearly as amazing to see an engineer shrug as it is to see a light bulb shrug, the effects are much the same.

Once upon a time, there was a large company called “Red.” Red Inc. had a team of salesmen who were, it seems, not producing the necessary volume of sales. While this may have gone a long way toward explaining the name of the company, it was not exactly a viable long-term strategy.

One day, the VP of Sales decided that the problem was clearly that the salesmen were not calling enough potential clients. They were wasting their time. They needed to be more efficient with their calls.

Much effort was spent focusing on the calling habits of the salesmen. They were given scripts. They were forced to practice making calls with various managers listening in and rating them on their performance on these practice calls. Those salesmen who demonstrated too great, or at least too obvious, a reluctance to make calls were dismissed. Those who questioned whether this was the right way to approach the problem either learned quickly to shut up or were also dismissed.

The sales team became very efficient at making calls. Sales did not increase. The remaining salesmen shrugged.

It turns out that even the best salesmen are reluctant to make calls. The problem was not with making the calls. The problem was with projecting the necessary confidence and optimism to attract and hold the interest of the client. Clients, it seems, are not all that likely to buy from salesmen who do not appear enthusiastic and confident in what they are selling. It also helps to know how to close the deal.

In each of these situations, a goal was set, a metric for success was defined, and that metric became the sole determinant of progress. Goals are extremely powerful tools: the best thing about them is that you accomplish them. Unfortunately, sometimes the worst thing about goals is that you accomplish them. In each of these examples, they accomplished their goals. A dead light bulb is extremely efficient, but not useful. Similar observations can be made about the server and the sales team.

Before leaping into setting a goal, especially a goal to solve a problem, it helps to understand the actual problem and to understand what the actual symptoms are. At Red, they assumed that an unwillingness or inability to make calls was the cause of the low sales and set their goals accordingly. We’ll never know how many top salesmen they dismissed because they didn’t realize that even the best salesmen suffer from call reluctance. Rather than create useful goals, they fixated on a symptom. That did not, however, actually change anything.

At both Soak and Red, the respective VPs stated that they were trying to solve the problems their companies were facing as rapidly and effectively as possible. They were setting goals. They were Taking Action! Taking action is certainly helpful, but it is even more helpful to be taking the correct action. Since it’s not always possible to determine just what the correct action is, it becomes even more critical to listen to the feedback and questions from the people who are charged with actually executing the action. The engineers and the salesmen knew that something was wrong, but no one was willing to listen to them. Remember, a key aspect of successful goal setting is understanding the feedback you’re getting.

I realize that many of you reading this are probably chuckling to yourselves and thinking that this scenario could never happen at your companies. The folks at Soak and Red said the same before, during, and even after it happened to them. The light bulb had no comment.

Setting a goal, for example, to be more efficient, seems like it makes sense and certainly feels good. However, it pays to determine if that goal is actually going to get you what you want. Otherwise, you may just end up with a dead light bulb.

© 2011 – 2014, Stephen Balzac. All rights reserved.

Share this article:Share on LinkedInShare on RedditTweet about this on TwitterShare on Google+Share on FacebookEmail this to someone