Tuesday, May 10, 2011

Why isn't Agile "winning"? (part 2)

In Part 1 of this post I made two assertions:
  1. Our capitalistic business community is quick to recognize and adopt practices that have a clear impact on the bottom line.
  2. That same community is not so great at adopting complex practices, and as a result settle for mimicry of the process rather than true adoption.
Is this really true? Good question. In the first part of this posting we discussed a few examples of how companies failed to successfully adopt complex practices. Now let's look at an example of how a company properly declined to try to replicate a successful complex practice.

In 2007 FastCompany published this article about a GE jet engine manufacturing plant in Durham, NC. It's definitely worth a full read, but in the interests of brevity I will summarize the salient points:
  1. This plant successfully competes against other engine manufacturing plants both inside and outside of GE, including off-shore manufacturers that have labor costs that are pennies on the dollar compared to US labor costs.
  2. Only 1/4 of engines from this plant ever have a single identified defect (almost always cosmetic). 3/4 of the engines produced here are defect-free.
  3. There is no management layers. There is a Plant Manager, and the manufacturing teams. Within the manufacturing teams there is only one worker classification
In 2008 this plant made a decision to start manufacturing a new engine that was in increasing demand and already being manufactured by other GE plants. Within 9 weeks of that decision the plant shipped their first engine, at a cost approximately 12% less than what other GE plants that have been manufacturing that engine for years were delivering it at.

In the world of manufacturing a 12% cost reduction is a phenomenal achievement. To achieve that with the very first engine shipped is unheard of.

So why isn't GE doing everything in their power to replicate how the Durham plant operates with all of their other manufacturing plants? Is the senior management in GE too incompetent to recognize that a 12% decrease in manufacturing costs is a good thing? Not likely.

At the end of the article the Plant Manager of a different manufacturing plant is quoted as saying "I think what they have discovered in Durham is the value of the human being. Here we have people that turn wrenches. In Durham they have people that think.". 

It's clear that GE understands why and how the Durham facility does what it does. But it is equally clear that GE isn't rushing to replicate the Durham model elsewhere. Why? Because that model is too complex to replicate. You cannot achieve the same results by mimicking the process. You also have to replicate the people. Remove the people that make the process work and you have a disaster in the making.

GE is smart enough to know that a manufacturing disaster leads directly to massive real world costs. Have a few jet engines fail in flight due to manufacturing defects and GE is quickly out of the jet engine business. But what about other practices that do not translate so clearly to the bottom line?

Let's look at social media engagement. There is a lot of pressure on business today to "get into" the social networking space lest they be left behind. Engaging social networking is a tricky business because you cannot control the conversation, you can only participate in it. Last year Nestle decided that they were going to squash the practice of Facebook users using an altered Nestle logo as their profile picture. Their subsequent handling of the conversation has become legendary in the social networking sphere as a perfect example of how not to interact with your community.

How did Nestle find itself in this position in the first place? Complexity. They "knew" that they needed to be engaging in social networking lest they be left behind by their competitors. Unfortunately they did not properly understand the complexities of engagement with customers via social networking until too late. Fear of losing ground in the marketplace led to imperfect mimicry of the complex practice of engaging social networking. Nestle would have been better off completely avoiding any social networking engagement.

Agile is no question a complex practice. For every story you read about how Agile has become a mainstream practice there is another that laments the fact that Agile adoptions are mimicry at best and not true adoptions. William Pietri (whom I have joyously crossed swords with in other forums) has a great post about this lack of true adoption. The telling quote from the article:

"I think there is a vast army of supposedly Agile teams and companies that have adopted the look and the lingo while totally missing the point."

As we've seen, mimicry of a practice is not the practice. Without the actual practice you cannot reap the benefits. If we cannot clearly show the benefits of Agile within our own software industry are we going to convince the Boardroom to flock to Agile just because we say so?

Of course! If ignoring actual results in favor of a convincing sales pitch is good enough for politicians, it's certainly good enough for us. Vote Agile! Change you can believe in (this time we mean it)!

Seriously, I believe that there are two paths to establishing Agile as a valid winning business strategy. Let's take a quick look at these paths:
  1. Clean up our act. We've done a great job of selling the software industry on the fact that attending a three-day certification seminar where the only critical evaluation of your Agile skills takes the form of a one-page written test, the answers for which are usually dictated by the course facilitator. If we're going to sell Agile, we need a proper means of effectively measuring Agile. When someone comes to me claiming to be a Certified Scrum Master, I want to see proof that they have the ability to eliminate obstacles or conduct stakeholder feature negotiations effectively, not proof that they were capable of keeping their butt in a chair for three days. If an Agile coach shows up and promises full Agile adoption in anything less than a year of ongoing engagement with the team I want to see how well their previous clients have adopted Agile.
  2. Beat them at their own game. There is no question that there are teams that are truly Agile, that deliver value frequently and consistently. There are enough of them now in the software industry that they have already become the "shining example" of success that the rest of the industry is attempting to duplicate. There was a tipping point when there were enough of these successful teams in the software industry talking about their success to allow the concept of Agile to become mainstream within the software industry. There are businesses out there right now that are winning at business in an Agile manner whether they know it or not. If we can properly market these businesses as Agile, how many more are needed to reach our tipping point?
Sadly, although I am passionate about the first option, I am also a realist enough to realize that there is no way we'll convince the hordes of coaches, trainers and even certification bodies out there to clean up their act. There's far too much money at stake for them to do anything different. And unless their customers realize that they aren't getting what they are paying for, there is no impetus for change.

It's the second option that I think has the best chance for success. There is a part of me that feels like giving up on the dream of true mainstream adoption of Agile (for software or business in general) is a personal failing on my part. But the reality is that Agile is a complex practice, and nothing we can do will reduce that complexity. Complex practices are doomed to be understood and exploited by only a small percentage of businesses and at best mimicked imperfectly by the majority.

So if we want to see Agile win acceptance outside of the team room we need to take an active hand in creating the success stories that lead to the mainstream acceptance tipping point. If we're happy with what we are achieving in the team room it's time to take those successes and bring them into the board room so that other parts of the business can start to realize the value of Agile as a means of conducting business.