Finish your things
Starting things is exciting. There's nothing like the feeling of potential that starting a new project brings. The world is your oyster and you have an opportunity to make something great. This time will be different. All those other projects you started were just practice runs. This is the real deal!
For engineers and product managers alike, starting a new thing is a nice excuse to retreat into the comfort zone. There are certain tasks which have to happen at the start of every project. Setting up test pipelines, creating the skeleton of the project, creating a lean canvas, figuring out all the cool features you will build.
It can be cathartic to work on these simple things. In this work you're insulated from the messiness of the real world. You're doing something more noble. You're laying a strong foundation for all the impactful work that will happen in the future.
One thing these early tasks have in common is that they happen far from the eventual customers. They take place in a bubble-wrapped world where nobody tells you you're wrong.
Finishing things is not like that. By necessity, finishing requires that your ideas, code and design crash into the chaos of the real world. Finishing can be vulnerable. It's easy to pretend that your product or feature is perfect as long as you never get it in front of the customers. The customers will see your shame. They don't care that you spend 3 weeks perfecting your test pipeline. They just want to know if their problem is solved.
You may have been working on a particular product or feature so long that you're sick of looking at it. You may be so used to it that you're not learning anything by sticking with it. You may be pushing for weeks to finally roll something out and every time you do you realize you're going to break another edge-case user who is doing something weird with the product. All of this sucks.
In these moments, it's tempting to give up finishing things and go start something new. Don't give in to this temptation. Remember that it is finishing things that brings value to the customer. You can start 10 different things a month and generate zero value unless you actually finish one and ship it.
A good rule of thumb is to always run towards the pain. Prioritize finishing over starting. You'll know you're doing the right thing because it will be difficult. It feels like you're going fast when you're constantly starting new things but it is finishing which generates outcomes. Finish.
Thanks to John Cutler who originally turned me on to this idea.