Read on Twitter
1/ I seem to have a habit of taking on projects that are often lofty dream ideas that had been tried and tried before and littered with bodies of failed previous attempts
2/ I haven't always been successful at them, but in my career my success rate has gone up significantly over time
3/ The projects are often ones that require buy in from a lot of different "owners" and worse there typically there is no clear team/person that owns the success
4/ One of those was back at Heroku the ability to attach a database to multiple apps (known as attachable resources to us at the time)
5/ Thinking back some on why I've become more successful on those things at different companies, a few learnings:
6/ First, never try to take on that project as your first project. As the new kid on the team it's the easiest way to lose credibility because you're 'discarding all that came before'
7/ Second, get buy-in and consensus gradually. The best way of doing this is asking people what they want, not prescribe what you want to do
8/ This is a great thread on the buy-in process ( https://twitter.com/jacobian/status/1351217224476983297)
9/ As you're getting buy-in, you gradually want to lay out what the overall goals are. On failed projects often the goals then vs. now have changed and people are misaligned.
10/ (This one was a big learning for me). A lot of people lay out the entire plan from the start. I appreciate this personally. I can think long term, but not get analysis paralysis
11/ Not everyone works this way. When I inherited a new team of engineers it crazy stressed them out, they wanted me to be super prescriptive.
12/ 4 months into this small building block, then that one, they paused and looked back and said you saw this coming all along.
13/ You fully need to have the big picture, but don't assume your team does. Sometimes they do and that's great, sometimes it causes more disruption than clarity.
14/ Perhaps most important, be patient, the more you force something the more resistance you'll find. Progress is a thing and perfection can absolutely be the enemy of good on these types of projects.
