We Didn’t Stop to Ask If We Should: Understanding Build vs. Buy
summary
Retrospective of build vs buy shows concrete numbers and failures: a custom security system cost two engineers’ time, roughly $100k upfront plus about $7,800/year maintenance and a $4,800/year loss versus a $3,000/year cloud alternative, and the team ends with Dilophosaurus factor = 1 (single engineer critical). Use the Baer Metrics calculator to compare costs, or build an error monitor with Elasticsearch + Kibana (initial ~$7.6k, $3.2k/yr) versus a hosted option like TrackJS at $2.7k/yr. Ask impact of failure, uniqueness, sustainability, and opportunity cost.
AI-written summary · about
chapters (28)
- 0:00Intro
- 0:24An Uncomfortably Close Look
- 2:09Build is Unique
- 2:32Build is fun!
- 3:09Enable Ninja Developers
- 3:39Digging for the Details
- 4:33"Boot up the door locks!" Dr. Alan Grant
- 4:50Bus Factor
- 5:08Dilophosaurus Factor
- 5:50Maintenance Time
- 6:34Handling Outages
- 9:30"We have all the problems of a major theme park and a major zoo" Ray Arnold
- 11:06Maturity is overrated.
- 12:13Buy is Fast
- 13:45Project Manager with a shotgun
- 14:22Not-Invented-Here Syndrome
- 16:33System Configurability
- 17:30Vendor Lock-In
- 18:28Sticky
- 23:12Mud on our Face
- 23:48Hello Newman!
- 24:04The Questions We Should Have Asked
- 24:22Impact of Failure
- 26:40Bias for Building
- 27:24Don't Make it Policy
- 27:56Error Monitoring an example
- 29:11Alerting
- 32:09Opportunity Cost
related talks
47mBuild vs Buy: Software Systems at Jurassic Park
30mBuild vs. Buy - a strategic decision for systems and people
46mBuild, buy, or use open source? All your answers in one chart
11mMaking better build vs buy decisions
30mHow we chose between building vs buying software
25mMaking smart investments: A framework for maximizing your ROI in technical decisions
16mBuild vs. Buy
29mMaking the Business Case for DevOps