How to sell a big refactor or rewrite to the business?
summary
Argues that big rewrites succeed only when they deliver measurable user value incrementally or deliberately create new competition. Cites Prezi’s ActionScript→JavaScript rebuild, VS Code rewritten on Node to win web devs, a monitoring-driven fix that cut catastrophic error costs from thousands to hundreds, and a daily import sped 25 hours to 5 minutes by a single SQL query rolled out behind a service switch. Uses monitor-driven prioritization and migrate by exposing APIs as concrete tactics and shows these projects worked because customers or metrics improved during the work.
AI-written summary · about
chapters (11)
- 0:00Start
- 3:24The design stamina hypothesis
- 8:10Lean Poker
- 9:37The risk of rewrite
- 11:54The abandoned platform
- 13:23Catastrophic failure
- 15:34Story of the slow import
- 17:42The talk from yesterday
- 18:52Becoming a platform
- 26:43Create your own competition
- 28:25Incremental replace with user value
related talks
01h 17mRefactoring the ROI of Software
43mHow to Innovate in a Legacy
36mRewrite, Rework or Refactor
51mA Teacher, an Economist and a Developer Walk Into a Bar
04mInterview with Swizec Teller
23mRefactoring Go in Large Codebases
50mRefactoring Is Not Just Clickbait
40mLegacy Evolution – The Innovation Opportunity!