Is always higher than you think.
We have been working on these hospital deployments at work at full steam for 8 months. Our first timeline was end-user training and go live in 4-5 months.
This happens all the time. I would venture to guess that more than half of all major IT projects fall behind schedule. It's human nature. Most people make aggressive planning assumptions most of the time.
I've found a mental framework that really helps me think about major planning decisions to avoid this pitfall: What is the cost of being wrong? And how does will that cost vary the more wrong I am? 2 stark examples:
If he had considered that cars and people move randomly on a public road, he might have chosen a better place to ride his bike backwards.
Nokia is perhaps the most glaring example of a modern company that failed to really think about the cost of being wrong, twice. In 2007, when the iPhone launched, Nokia was king of the world. They were incredibly wrong about the future of mobile computing. And when they did finally realize that in 2011, they chose the wrong partner. Now, it seems that Nokia's only chance for survival is acquisition by the company that banked on to turn themselves around: Microsoft.
We decided today to stick to our plan to go live on March 1st. I was against it, though things are looking better than they did 24 hours ago. A few requirements have been scaled back. As we finalize our hospital deployments at work today, I can't but think to myself, what is the cost of being wrong with regards to functionality? data conversions? training and go-live methodology? support?
The problem of course is that these are all very difficult questions to accurately assess. Even still, forcing myself to think in the "what if I'm wrong" framework helps make better business decisions.