Unreasonable Estimates and Improbable Goals
summary
Advice: clarify the exact requirement and why it’s wanted, break work into database Rails and front‑end pieces, list risks (IE8 quirks, denormalizing tables, large data loads), give a range not a single number, and quantify uncertainty with a best/worst spread. Call out hidden costs—operational (Resque/Redis deployment and monitoring), support, and opportunity—and negotiate features not time. Update estimates as learning occurs and resist estimation bargaining and deadline hardening.
AI-written summary · about
chapters (20)
- 0:00Intro
- 0:25Adam Sanderson
- 1:01Liquid Planner
- 2:08Clarify
- 4:31Estimating
- 7:19Seriously
- 7:33Quantify Uncertainty
- 9:14Deal With Uncertainty
- 11:29Estimation Bargaining
- 13:16Complexity Costs
- 14:23Operational Costs
- 15:09Support Costs
- 16:15Opportunity Costs
- 16:55Weigh the Costs
- 17:25Ego Manipulation
- 18:53Deadlines
- 19:51Dealing with it
- 23:55Overtime
- 26:09Deadline Hardening
- 28:13Any Questions?
related talks
29mSoftware estimation – embracing nuance and controlled chaos
47mCreating efficient, accurate, software estimations
41mSo you think you can estimate? The lost art of useful estimates
45mFrontend Estimation: My Story of Delivering Tasks on Time
11mThe key to solving the conundrum of software estimation
01h 45mEstimates or NoEstimates
54mEstimates – the Enfant Terrible of Software Development
46mEstimates and complexity: a tale in 42 and a half minutes