The Terrible Twos. Every parent’s nightmare. The age when your child learns the word “no” – and uses every available opportunity to use it.
If we were objective, we’d realize that the milestone of “no!” is really a dream come true. Along with “do it myself!”, it’s a sign of your child’s developing independence. It’s a critical stage on the way to a healthy childhood and adulthood.
But it’s uncomfortable – both practically (when you just need to get their shoes on and get them out the door) and emotionally (because your baby doesn’t seem to need you anymore).
Sometimes Scrum Masters feel like parents of two-year olds (or maybe teenagers):
The other week, one of the Scrum Masters I mentor confided in me. He saw his team becoming more and more autonomous, able to organize effectively even without his guidance. He knew that was the direction that Agile enabled and encouraged, but… “Haim, be reasonable! You and your Agile – you can’t run development like this!”
I smiled. “My Agile and I are thrilled,” I told him. “You’re very lucky. I guess you don’t realize how hard it is to enable a team to manage themselves. It can take months, years… and sometimes it never happens at all! You should be throwing a party!”
I’m not sure if he was convinced. But I was.
Let’s take a look at two stories (out of many!) of teams whose independence and self-organization brought their companies huge benefits.
From ‘Suck’ to Spectacular:
Eric Schmidt, Google’s chairman and former CEO, tells the following story:
When Adwords first launched as a product, Larry Page (one of Google’s founders) searched for the term “Kawasaki H1B” (a vintage motorcycle). He got back ads from lawyers offering to help would-be US immigrants get H1-B visas, but no motorcycle ads. Then he searched for “cave paintings”. The Adwords results featured art galleries – none of which sold cave paintings.
This was a serious problem. Irrelevant ads would get no clicks – leaving Google Adwords with no revenue.
A traditional manager would have convened an emergency staff meeting to decide upon a special forum to analyze the problem. Two weeks later the forum would present an analysis and a dedicated team would be formed to work day and night until they solved the problem.
What did Larry do?
Larry printed the results out and highlighted the irrelevant ads with the written explanation, “THESE ADS SUCK.” He stuck the paper on the bulletin board in the staff kitchen and went home for the weekend.
Three days later, Larry got an email from a group of five Google developers, none of whom were part of the Adwords team. They had seen the page, understood the issue and spent the weekend working on a prototype. Attached to the email was their analysis and suggested solution.
Their voluntary, ad-hoc solution forms the basis of the Google Adwords algorithm until this day.
What happened there?
The solution came from a group of talented individuals, connected to the interests of the company, who independently decided to tackle the issue. No one asked, demanded, supervised or managed them.
They saw a need, a challenge, and in an entirely autonomous manner, they solved the problem as quickly and effectively as possible.
Eh, you might say. Google is Google.
I beg to differ.
PL1 to C Gets an A++
(Not so) long ago, I was a young, promising team leader at Elbit Systems. Something in me instinctively balked at expressing authority, and so I intuitively directed my team to become responsible, independent and self-organizing.
One day, our department head called me in a panic. We needed to convert all the central code of the secret system we were working on from PL1 to C. On the double.
Let’s see, I thought. That’s going to be a bit of a challenge. We need to plan the project, check the new design, code and check the entire system. That should take about two years to complete. And we have only a few months until we’re supposed to release the updated system. Hmm… can someone say “Mission Impossible”?
It was a Thursday. At our regular Thursday afternoon Happy Hour, I presented the challenge to my team.
“It’s obvious we’re not going to be able to accomplish this using standard practices,” I concluded. “But I know all of you and your abilities, and I’m confident that this challenge will inspire you to find creative solutions.”
Monday morning as we arrived at work, Sharon and Oren, two of my team members, approached me.
“We have good news for you and bad news for you,” Oren began. “The good news is that over the weekend we thought of a solution to the PL1 to C conversion issue and we wrote a program that can convert the code automatically. The bad news is that the code is only 80% accurate and the rest will have to be done manually.”
My response? “I should always have bad news like that! You guys are amazing – show me the code!”
I didn’t demand, direct, request or assign ANYTHING to ANYBODY. I simply presented the problem, and responsible, committed individuals rose to the challenge and created an effective and efficient solution. The project was completed successfully long before the deadline.
About a year later, I moved to another team. I discovered that they, too, had received a similar directive to convert large sections of code from PL1 to C. They had outsourced the conversion project to an American company, under a contract stating they were to receive the final code in C within 6 months.
Two years later, the code had still not been compiled…
So, my dear Scrum Master, I have bad news for you and good news for you.
The bad news is that if you do a good job, your team will become more independent and need you less.
The good news is that they might accomplish things you never dreamed were possible.