Tuesday, November 06, 2007
Enterprise Architecture and the Titanic Effect
The thought that disaster is impossible often leads to an unthinkable disaster...
Turning the large enterprise in the direction it should head is a herculean effort. I am of the belief that the best way to accomplish this goal is for enterprise architects to focus on the creation of valuable working software. We tend to spend way too much time on working up the ladder while ignoring what it takes to create great software and the folks who from an organization chart perspective are below us.
What if we were to noodle ways to make ourselves more efficient at the lower levels? While I asked Why have process? maybe the question should have been what should enterprise architects acknowledge in their own minds when it comes to process. The analogy of manufacturing is the latest meme to invade the mindset of corporate America, so maybe the best way to uncover bad EA thinking is to think about software development as a plant.
Now, if we were to analyze each phrase, you may walk away with a couple of action items. Consider that if speed of the bottleneck is the speed of the plant then maybe you should have an initiative to figure out where the bottlenecks are. If it is your governance and folks are stepping in it, then maybe it needs to be removed. If Indian outsourcing slows down delivery of software then maybe you need to bring it back inhouse. If we focus on Dont run for individual station efficiency then we may acknowledge that we spend way too much time selling up the foodchain and need to spend more time interacting with the masses in our organization whom actually make things happen. We may even acknowledge while transparency is important, that we shouldn't sacrifice productivity for it.
Have you noticed that since IT outsourcing became popular, most enterprises haven't actually focused on making software development more productive? In fact, I suspect that most have made it more inefficient. The funny thing is that Indian outsourcing firms also never take steps to increase productivity as it would have the side effect of less hiring. Someone has to keep this process honest...
| | View blog reactionsTurning the large enterprise in the direction it should head is a herculean effort. I am of the belief that the best way to accomplish this goal is for enterprise architects to focus on the creation of valuable working software. We tend to spend way too much time on working up the ladder while ignoring what it takes to create great software and the folks who from an organization chart perspective are below us.
What if we were to noodle ways to make ourselves more efficient at the lower levels? While I asked Why have process? maybe the question should have been what should enterprise architects acknowledge in their own minds when it comes to process. The analogy of manufacturing is the latest meme to invade the mindset of corporate America, so maybe the best way to uncover bad EA thinking is to think about software development as a plant.
- The speed of the bottleneck is the speed of the plant
- Dont run for individual station efficiency
- Don't allow yourself to be less efficient at non-bottleneck stations
- Excess inventory is a liability not an asset
Now, if we were to analyze each phrase, you may walk away with a couple of action items. Consider that if speed of the bottleneck is the speed of the plant then maybe you should have an initiative to figure out where the bottlenecks are. If it is your governance and folks are stepping in it, then maybe it needs to be removed. If Indian outsourcing slows down delivery of software then maybe you need to bring it back inhouse. If we focus on Dont run for individual station efficiency then we may acknowledge that we spend way too much time selling up the foodchain and need to spend more time interacting with the masses in our organization whom actually make things happen. We may even acknowledge while transparency is important, that we shouldn't sacrifice productivity for it.
Have you noticed that since IT outsourcing became popular, most enterprises haven't actually focused on making software development more productive? In fact, I suspect that most have made it more inefficient. The funny thing is that Indian outsourcing firms also never take steps to increase productivity as it would have the side effect of less hiring. Someone has to keep this process honest...
