Ever had to talk yourself more or less completely out of a project?

Question

I recently had a project that almost failed miserably. To me it's a complete failure but somehow it passed, shouldn't have, just did. There was a lot of good will on all sides and the blame was shared equally. But now that I put the project files into "cold storage" I wonder what will happen in "the worst-case scenario". You are on a sinking ship, the complexity is exploding because there is a (not so) tiny hole in the specification, the customer is totally flipping out, it simply can't be done period

Ever had to talk yourself completely out of a project midways? How did this work out?
What was your reaction and what did you do (or intended to do) to avoid such things in the future? Like "From now on it's scrum or bust", "Darn, I will never do xyz again", "hm, does my insurance cover that" or maybe even "shit happens, stop whining and get over it"?

Answer 1

We had a project where one requirement was a guaranteed response time. After a few months into development we couldn't keep up to that promise. And what's worse we kept the problem under the hood way too long. The presentation went like: This is the last milestone and by the way we missed that response time thingy. Complete disaster. I had no part in the following negotiations but from all I've heard it was ugly. There were technical issues and design flaws that had to be addressed without doubt. But in my opinion the most crucial consequences were: We put way more emphasis on status reports now. Not in a micro-managing way, but at least more concise. Problems have to bubble up a lot faster. And internal as well as external presentation cycles now are a lot shorter than they were then. That alone wouldn't have solved the problem of course. But we could have put more, combined afford in time into the issue. And the client might not have reacted as strong as he did. I think you can't get the probability of failure to zero percent, so yes "shit happens, get over it". But you can work on getting it as low as possible. And you can work on what you might call the magnitude of failure. Therefore: do status checks early, address problems early, don't let fissures become hull breaches.

Answer 2

I quit on a project midway once, I can't say I was fleeing problems, it was mostly that another project interested me better. I went to see my boss, explained to him how I could contribute to the other project while I wasn't very comfortable with my current one. Be honest with your boss, this is the best way to end up where you will truly be happy.

A thing I have done many time is being force to move to a project halfway into trouble. Working like crazy because of the mistakes of other is not the best thing for motivation. The worse, the better you are at saving a project like this, the more prone you are to be assigned to this again because you are "so good at this".