Goal setting for productive engineers

The post argues that engineering work is too interruption-heavy for clean long-term plans to survive unchanged. It uses hyperbolic discounting and the motivation equation to explain why big goals lose force, then contrasts rigid SMART-style goals with small experiments that build momentum and self-trust.

Reading notes

#

• Engineering work is interrupted by incidents, on-call work, dependency changes, and priority flips, so planning assumes more stability than reality provides.
• Yearly goals can start strongly and then die under work chaos and life chaos, which also makes later goals harder to trust.
• Hyperbolic discounting helps explain why distant rewards lose against Slack messages and new requests.
• The motivation equation is made of expectancy, value, impulsiveness, and delay.
• Expectancy is the belief that the work can be done in the real environment.
• Value is the meaning or enjoyment received now and later.
• Impulsiveness is the pull toward distractions like doomscrolling or answering Slack.
• Delay is how far away the reward is.
• Big goals often fail because delay is long and expectancy is low.
• SMART and CSI goals work when the environment is clear and the path is known.
• Top-down goals fit situations with a clear outcome, such as finishing a course and using it for a feature by the end of the month.
• Top-down goals fail when they become rigid contracts, ignore context changes, or copy goals that do not fit the team.
• Tiny experiments are small, repeatable actions run for a short period to test what is worth doing.
• Tiny experiments focus on input rather than output and can create more information instead of rigidity.
• Simple daily commitments can build a chain and rebuild trust, such as writing for fifteen minutes each weekday.
• Tiny experiments work well when the future is ambiguous or when a SMART goal is hard to write.
• A short experiment can be easier to commit to than aiming for promotion.
• Tiny experiments raise expectancy because the goal is small enough to hit even on a bad day.
• They raise value when the experiment feels interesting or fun.
• They reduce delay because they usually last one to four weeks.
• They reduce impulsiveness because ten minutes is easy to start.
• Tiny experiments fail when they become too many micro goals or never get turned into outcomes.