Scene: Surprise meeting with the project owner 0-3 days before the go-live date
“Hey team, the business and I have decided to postpone the project release by n=1-3 months because [they aren’t ready for it / it isn’t finished /regulatory reasons]. And since we have some extra time now, we can tie up all the loose ends on this project (i.e., ‘we’ve added n+1 months worth of backlog items to the MVP’).”
I’m still a greenish dev, so maybe this is normal, but I’ve had the same story going on for over a year now, and it’s really starting to burn me out. In the beginning, I was optimistic. Now I just hope for the project to fail, or me to get off somehow, but this thing just won’t die.
Anyone with experience on similar projects able to share words of advice? Do they ever end up working out? Seems there’s a death spiral, since we are always rushing to a deadline, forgoing tests and quality but never cleaning up our mess because we’re already behind. Yet I somehow feel like I’m the crazy one for thinking this 6-month “quick” side project turned 2+ year half-rewrite will have trouble meeting it’s Nth deadline.
This sounds like the old waterfall approach to development.
Design the whole system, create the whole system, test the whole system.
The problem with this approach is that requirements almost always change or scope creeps in those timeframes.
Now most companies are bad at agile. But even moving slightly towards it is better than nothing.
This will continue to happen until the project management changes unfortunately.
If the system can’t be worked on in small independent chunks this is basically guaranteed to happen.
We do agile very loosely. But we have a two week sprint and at the end we, hopefully, have the features we decided on done and deployed. Then we can get feedback and add the required changes to the backlog for a future sprint.
This way you get feedback a lot quicker and as you pick work every two weeks you can keep things moving forward.
Chances are the company won’t change so if it bothers you looking for another job may be a good shout.
It’s very common. I’d bet most software projects still fail. I once met a guy who’d been a gamedev for over 10 years at some big companies, and every game he ever worked on got cancelled.
Sometimes these long, poorly managed projects do succeed though. Usually because they launch a beta or get publicly scheduled for release, and the sudden reality check causes someone senior to trim the scope down dramatically.
It might be worth sticking around if you think you’re learning things, but impose some personal limits. Don’t kill yourself trying to meet some idiot’s impossible schedule. Work your contracted hours and no more. Be blunt and unashamed about how long tasks really take. Share your concerns when the month’s schedule looks unrealistic, referring back to previous tasks that overran. This is how you learn to one day become a lead who doesn’t run shitty projects like the one you’re on.