Have you ever been stuck on an Andy Dufresne project?
You know the one. You’re trapped because of something someone else did, chipping away at legacy structures with just a rock hammer. And you might even have your own Rita Hayworth poster to hide the refactoring of this debt.
Sometimes that’s agency life. The promise of a rewrite (and your own boat in Mexico) if you can just chip away slowly and then crawl through half a mile of shit. It can be achingly slow progress, trying to get buy-in for your freedom, when the people with the power just want to keep you where you are. You’re trapped because of what someone else did.
And if you stay in that code too long, you’ll get institutionalised and won’t be able to see a better way to fix this problem.
How to keep making progress
Firstly, it’s much easier to survive a project like this if you have the right team beside you. People who can accept the faults and accept they’re temporary. You need a team of gardeners, not innovators, who can dig deep and understand what’s actually happening, where the rocks and the edge cases are. And who’ll take a break to listen to music or otherwise relax and take stock and look up.
And you need to understand the progress will be slow, but you can put guard rails in place so that each bit of progress you make has tests and is better than you found it. And sometimes you have to shake the refactorings out as pebbles amongst the other work because management and the clients don’t want to see it.
Never stop asking for freedom to make bigger changes, but always pair it with trust. They’ll listen more if you respect their budgets and their time, and you can explain what’s easier and what’s harder.
The Great Escape
If you do get the chance to rewrite, there may well be external factors at play. Use them. Digital transformation, going mobile, getting AI-ready. If you can tap into the strategy and find ways to align what you want with that, it’s going to be much easier to get approval. Be honest with the proposal, but always keep an eye out for those opportunities.
Look after Red
The only way to get the job done and get where you want to go is to take the important people with you. The ones who can get you a better hammer, or a new cover for the next bit of refactoring. Honesty will always get you further, so long as you know who to trust. Make it easy for them, so they can come with you on the journey.
See you on the beach.
Top comments (1)
The project-health lesson here is strong. Some projects do not fail because the team lacks effort; they fail because the surrounding system keeps making the rational local move and the irrational global move the same thing. Naming that trap is often the first useful intervention.