The good life algorithm: iterative improvement over grand leaps

Executive overview

Most people try to improve their lives by picking a grand goal and leaping toward it — but this moves only one dimension of a complex, multi-dimensional life, often making other things worse. Jim Collins' daily tracking practice — rating each day on a -2 to +2 scale and noting what you did — offers a better model: make small, data-driven adjustments that compound over time.

The result is a simplex-style algorithm for life design: you don't know the destination in advance, but each step moves you closer.

Small, evidence-based iterative changes outperform gut-driven grand leaps for reaching a genuinely deep life.

The grand goal strategy and why it fails

• Picking one desirable property and making a big leap in that direction ignores all other dimensions of your life
• You might get more of what you aimed for while inadvertently making something else much worse
• The full landscape between where you are and where you want to be is unknown — there are traps, local optima, and obstacles you can't see from a distance
• Grand changes are high-stakes; if the direction was wrong, the cost is large

Jim Collins' good life algorithm

• Every day: one line describing what you did, hours of deep work, and a subjective score: +2 (great), +1 (good), 0 (meh), -1 (negative), -2 (bad)
• The score must be entered in real time — you cannot reconstruct how a day felt five days later
• Over months and years, patterns emerge: what activities cluster around +2 days, what clusters around -2 days
• Use that data to make small, targeted adjustments — more of the +2 conditions, less of the -2 conditions
• The approach mirrors the simplex algorithm in operations research: find the optimum through iterative steps without knowing the destination in advance
• The outcome is often idiosyncratic — a life you never would have designed from scratch, but one that genuinely fits

Navigating negative news as a federal worker

• Recognise the psychological warfare dynamic: the goal is to induce chronic anxiety, not just inform
• Set two fixed check-in windows per day (e.g., 10 minutes); outside those windows, do not follow the story
• Urgent, actionable information will come through official channels (supervisor, agency head) — you don't need social media for that
• Constant monitoring is free delivery of anxiety; withholding that is the best available counter-move

Taking over a shallow workplace: three north stars

• North star 1 — context switching is productivity poison. Every cognitive shift drains a finite daily budget; minimise the number of switches, not just the friction of each task
• North star 2 — deep work on needle-moving output is the priority. Protect it above all else; make the case that anything reducing cognitive output is a cost the business pays
• North star 3 — no pseudo-productivity. Visible activity is not a proxy for value; results are. Track work on a shared board — one or two active cards per person at most, no hidden inboxes
• Daily standups exist to surface blockers and hand off information, then people go work — not to generate more coordination overhead
• Communicate the north stars explicitly and repeatedly; every change should be traceable back to one of them

Recovering from a health setback

• Your brain reads temporary rest as a permanent change and catastrophises; it is not a permanent change
• Take your initial estimate of how much rest you need, then double it — the instinct is always to under-rest
• Historical figures we regard as prolific had long fallow periods: Darwin seasick on the Beagle for months, Kennedy bedridden with Addison's disease for weeks at a time
• Pseudo-productivity frames any inactivity as disaster; true productivity is measured over years, not days
• Three to six months of reduced output does not derail a career or a life's work

Escaping pseudo-productivity from inside a pseudo-productive organisation

• Trade clarity for responsiveness. If colleagues always know the status and expected completion date of your work, their need for immediate replies drops sharply — most urgency is driven by uncertainty
• Limit concurrent workload. Every active project carries administrative overhead; too many active projects means most time is servicing tasks rather than completing them
• Separate "actively working on" from "waiting to work on" — be transparent about the queue and communicate only about what is active
• Use quotas for recurring commitments: a fixed number per week or quarter, with a polite but firm cutoff when the quota is full
• Become genuinely good at something the organisation needs; leverage follows quality, not compliance

Defining "enough" for autonomous or flexible roles

• If the employer is satisfied and paying a fair rate without deception, the market has already validated that the work is enough
• The 40-hour norm derives from 20th-century factory labour negotiations, not from any empirical study of knowledge work
• Separate the job into two parts: client-responsive work (handle well, then stop) and self-initiated work (a phantom part-time job, a skill-building project, or a company-internal initiative)
• Content creation is a viable self-initiated pursuit but is a genuinely competitive and difficult field — go in clear-eyed
• Guilt about surplus time is a pseudo-productivity reflex; redirect it toward lifestyle-centric planning with a specific target

Books read in February 2025

• Moral Ambition — Rutger Bregman: a call to direct talent toward genuinely useful work rather than high-status but low-impact careers
• Brotherhood of the Rose — David Morrell: classic Cold War thriller; Morrell is also the author of First Blood
• How Dante Can Save Your Life — Rod Dreher: memoir of using the Divine Comedy to navigate a personal crisis; Dante history woven in
• Buzzsaw — Jesse Doherty: inside account of the Washington Nationals' 2019 World Series season; better writing than expected
• The Sirens Call — Chris Hayes: smart cultural-critic take on attention and the attention economy; no prescriptive advice, but well-researched and clearly argued