← Writing

Recurring Decisions

The field guide catalogs the sensors. This page inverts it: every recurring decision a product company faces, with the loops that inform it stacked from fastest proxy to slowest truth. The cheap signal on top is usually the most gameable; the honest one at the bottom is the one you wait for.

Your stage

Build

Can we make the thing?

Is this change correct — and safe to merge?

The deepest ladder in the catalog: from feedback before you finish typing to the bug report that arrives weeks later. Every rung you add on the left saves a more expensive lesson on the right.

  1. 300msIDE inline feedback
    highpush
  2. 500msSyntax / parser
    highpush
  3. 2sType system
    highpush
  4. 3sHot reload / REPL
    medpush
  5. 30sLint / static analysis
    medpush
  6. 1minRuntime schema validation
    highpush
  7. 1min◆ Pair / mob programming
    medpush
  8. 1minSecrets detection
    highpush
  9. 2minUnit tests
    highpush
  10. 5minAI code review
    medpush
  11. 10minCoverage / mutation testing
    medpush
  12. 10minIntegration tests
    highpush
  13. 10minVisual regression testing
    medpush
  14. 30minE2E tests
    medpush
  15. 30minSAST / code security scan
    medpush
  16. 1hrError monitoring
    highpush
  17. 12hr◆ Code review
    medpush
  18. 1wk◆ Bug reports
    medpush

Is production healthy right now?

The always-on loops. Everything here is push and minutes-scale — if any of these are pull, you find out about outages from customers.

  1. 5minSynthetic monitoring / uptime checks
    medpush
  2. 15minAlerts / on-call pages
    highpush
  3. 1hrRuntime security detection
    medpush
  4. 1hrAPM / distributed tracing
    highpull
  5. 1hrMemory / resource utilization
    medpush
  6. 1hrError monitoring
    highpush

Can we keep shipping safely?

Not "is this deploy good" but "is our shipping process trustworthy": blast-radius control on the fast end, error budgets and failure rates as the trend.

  1. 30minCanary / progressive rollout
    highpush
  2. 1d◆ Chaos engineering / game days
    highpull
  3. 1wkSLO / error budget burn
    highpush
  4. 1moChange failure rate
    medpull

Could an adversary get in — and would we know?

The full security ladder: the flaw caught at the whiteboard, the leak caught at commit, the inherited CVE, the drifting cloud config, the live attack — down to the commissioned pen test and the unbounded bounty crowd. Vulnerability age is the loop that watches whether any of it actually gets fixed.

  1. 1minSecrets detection
    highpush
  2. 30minSAST / code security scan
    medpush
  3. 1hrRuntime security detection
    medpush
  4. 1d◆ Threat modeling
    medpull
  5. 1dSCA / dependency vulnerabilities
    medpush
  6. 1dCloud misconfiguration / CSPM
    highpush
  7. 1wkDAST / automated web scanning
    medpush
  8. 2wk◆ Bug bounty / VDP
    highpush
  9. 2wk◆ AI red-teaming / adversarial evals
    highpull
  10. 1moOpen vulnerability age / patch SLA
    highpull
  11. 3mo◆ Pen test / red team
    highpull

What broke — and what do we learn from it?

The failure-learning ladder, including the loop that runs before the failure exists: pre-mortems are post-mortems with better timing.

  1. 1d◆ Pre-mortems
    lowpull
  2. 1wk◆ Incident post-mortems
    highpull
  3. 1moMean time to restore
    medpull

Is it fast enough for real users?

Lab numbers in minutes, field truth in weeks. The lab loops are proxies — the 28-day field window is the number that ranks you and retains users.

  1. 3minBundle size / asset weight
    highpush
  2. 10minLighthouse / synthetic performance
    highpush
  3. 30minBenchmark tests
    medpush
  4. 1hrModel latency / time-to-first-token
    highpull
  5. 1wk◆ Load / stress testing
    medpull
  6. 1moCore Web Vitals (field / RUM)
    highpull
  7. 1moCapacity / scaling signals
    medpush

Is our delivery machine healthy?

Flow, not effort: how long work waits, how much is in flight, how often it ships, whether plans mean anything. The standup is the human sensor; the rest are the instruments.

  1. 1d◆ Daily standup
    medpush
  2. 1dWIP
    highpull
  3. 1dBuild & CI duration
    medpull
  4. 2d◆ App store review gate
    medpush
  5. 3dWork item age
    medpull
  6. 2wkTask cycle time
    medpull
  7. 2wkThroughput
    medpull
  8. 1moPR cycle time
    medpull
  9. 1moDeployment frequency
    medpull
  10. 1moLead time for changes
    medpull
  11. 1moSprint predictability (say:do ratio)
    lowpull
  12. 1moFlow efficiency
    medpull

Where should we invest in the codebase?

Three rungs of the same ladder: static metrics lie fast, hotspots tell the truth in months, the tech-debt trend settles it in years. Refactor where churn meets complexity, not where the linter complains.

  1. 10minCode complexity & coupling metrics
    medpush
  2. 1moHotspot analysis (churn × complexity)
    highpull
  3. 1moDependency freshness (libyear)
    highpull
  4. 1yr◆ Tech debt accumulation
    lowpull

Can we even build this?

Feasibility answered by building the smallest dangerous version, and by auditioning the technology before you marry it.

  1. 1d◆ Feasibility spike / throwaway prototype
    highpull
  2. 1wk◆ Technology / model evaluation
    medpull
  3. 1wk◆ Design / RFC review
    medpull

Can downstream consumers trust the data?

Data breaks silently — the dashboard renders either way. These loops exist to make staleness and shape-drift loud.

  1. 15minData quality tests
    highpush
  2. 1hrPipeline freshness / data SLAs
    highpush
  3. 1dData anomaly detection
    medpush

Is the AI feature actually good?

Quality no human can read at production scale: golden sets gate the change, judges watch the drift, users vote one thumb at a time, and the upstream model shifts under all of it.

  1. 30minEval suite / golden-set regression
    medpush
  2. 1hrToken / cost per request
    highpull
  3. 1hrLLM-as-judge scoring
    medpull
  4. 1dThumbs up/down on AI output
    lowpull
  5. 1wkModel & output drift
    medpull
  6. 2wk◆ AI red-teaming / adversarial evals
    highpull

Value

Is it worth making?

Should we build this at all?

The cheapest place to be wrong. Everything here runs before serious build cost: fake doors, interviews, and pricing research are how you lose a week instead of a quarter.

  1. 1wkFake door / prototype tests
    medpull
  2. 1wk◆ Weekly discovery interviews
    highpull
  3. 1wk◆ Feature request board / upvotes
    medpull
  4. 2wk◆ Kano survey
    medpull
  5. 2wk◆ Opportunity scoring (importance vs satisfaction)
    medpull
  6. 2wk◆ Willingness-to-pay research
    medpull

Can people actually use it?

From your own team's complaints to strangers' rage clicks. Dogfooding is fast but biased; five usability sessions find most of what matters; the field signals confirm at scale.

  1. 10minAccessibility checks
    medpush
  2. 1d◆ Dogfooding
    medpush
  3. 1dFirst-click / findability
    highpull
  4. 1dClicks / interaction cost
    medpull
  5. 1d◆ Task ease (SEQ / SUS)
    medpull
  6. 1dSession recordings / rage clicks
    medpull
  7. 1wk◆ Beta / early-access program
    highpull
  8. 1wk◆ Usability tests
    highpull
  9. 1wk◆ Task success rate
    highpull
  10. 1wk◆ Time on task
    medpull
  11. 1wkFunnel / drop-off analysis
    highpull

Did the thing we shipped actually work?

Stakeholder nods in the demo, causal proof in the experiment, and the only number that settles it: do people reach the value moment and come back?

  1. 1wkFeature adoption / activation
    medpull
  2. 1wkFunnel / drop-off analysis
    highpull
  3. 2wk◆ Sprint review / stakeholder demo
    medpush
  4. 2wkA/B experiments
    highpull

Do we have product-market fit — and are we keeping it?

The slowest, highest-stakes question in the value track. Sentiment loops lead, retention curves decide, and churn interviews explain the verdict after it's in.

  1. 1dAI synthesis of qualitative streams
    medpull
  2. 1wk◆ Public reviews (G2 / app stores / Reddit)
    medpull
  3. 1wk◆ Churn interviews / exit surveys
    highpull
  4. 2wkSean Ellis PMF survey
    medpull
  5. 1moNPS / CSAT
    lowpull
  6. 1moStickiness (DAU/MAU)
    medpull
  7. 1moTrial / freemium conversion
    highpull
  8. 3moRetention curves
    highpull
  9. 6moExpansion / referral revenue
    highpull

Are we priced right?

Assembled entirely from loops that live elsewhere — pricing is the decision most companies inform by accident. Willingness-to-pay asks before you commit; discounting reveals the truth during every deal; churn and lost deals deliver the verdict after.

  1. 1wk◆ Churn interviews / exit surveys
    highpull
  2. 2wk◆ Willingness-to-pay research
    medpull
  3. 1moTrial / freemium conversion
    highpull
  4. 1mo◆ Win/loss analysis
    highpull
  5. 3moPricing realization / discount rate
    highpull
  6. 1yr◆ Pricing power
    medpull

Is support scaling with the product?

Two different questions hiding in one queue: are we handling tickets well (speed, one-touch resolution, CSAT) and why do tickets exist at all (contact rate — the product-quality number wearing a support costume).

  1. 4hrTime to first response
    medpull
  2. 1dResolution time
    medpull
  3. 1wkTicket CSAT
    lowpull
  4. 1wk◆ Support tickets
    medpush
  5. 2wkFirst contact resolution
    medpull
  6. 1moContact rate / tickets per customer
    medpull

Reach

Can we get it to people?

Which channel deserves the next dollar?

The attribution ladder in full: last-click flatters, multi-touch negotiates, self-reported surprises, lift tests prove, and MMM allocates. Fast numbers spend the budget; slow numbers tell you whether they should have.

  1. 1dCPM / auction pressure
    medpull
  2. 1dLast-click attribution
    lowpull
  3. 1wk◆ Self-reported attribution
    medpull
  4. 1wkMulti-touch / data-driven attribution
    medpull
  5. 1wk◆ Influencer / sponsorship performance
    lowpull
  6. 1wk◆ Events / webinar performance
    medpull
  7. 1moROAS / channel CAC
    medpull
  8. 1moIncrementality / lift tests
    highpull
  9. 3mo◆ Partnership pipeline
    medpull
  10. 3moBlended CAC
    medpull
  11. 6moMarketing mix modeling
    lowpull

Does this message land with this audience?

Creative feedback at every speed: the algorithm answers in hours, the click in days, the reply in weeks. All of it is the same question — did the words earn the attention?

  1. 4hrOpen rate
    lowpull
  2. 4hrClick-through rate
    medpull
  3. 6hrViews / impressions
    lowpull
  4. 6hr◆ Comments / audience conversation
    medpull
  5. 6hr◆ Comments & DMs
    medpull
  6. 12hrCompletion / loop rate
    highpull
  7. 1dHook rate / swipe-aways
    medpull
  8. 1dAudience retention / watch time
    highpull
  9. 1dPaid ads CTR / CPA
    medpull
  10. 1dEngagement rate
    medpull
  11. 3dLanding conversion rate (CVR)
    medpull
  12. 3dContent engagement
    lowpull
  13. 3d◆ Cold outreach reply rates
    medpush
  14. 1wkQuality / relevance score
    medpull
  15. 2wkFrequency / creative fatigue
    medpull

Can we still reach the inbox?

Deliverability is a licence, not a metric — revoked quietly and restored slowly. Unsubscribes are the leading edge of losing it.

  1. 10minBounce rate
    highpull
  2. 15minDelivered / inbox placement
    highpull
  3. 1hrSpam / reject rate & sender reputation
    medpull
  4. 1dUnsubscribe / list churn
    highpull

Are we compounding an audience or renting one?

Rented attention stops when the spend stops. These loops measure the owned kind: followers, saves, community, unprompted recall — the assets that survive a budget cut.

  1. 1dShares & saves
    highpull
  2. 1dFollows-per-view
    medpull
  3. 1d◆ Community activity & sentiment
    medpull
  4. 2wk◆ PR / press coverage
    lowpull
  5. 1moFollower / subscriber growth
    medpull
  6. 1moReferral / K-factor
    medpull
  7. 1yr◆ Brand awareness / recall
    lowpull

Can buyers find us when they go looking?

The slowest channel to build and the cheapest to keep: keyword bets confirmed months later by rankings and traffic, decay quietly clawing it back, and answer engines rewriting the rules mid-game.

  1. 1d◆ Keyword research (difficulty vs. potential)
    medpull
  2. 1wkTechnical SEO / indexation
    highpull
  3. 1moKeyword rankings / SERP position
    medpull
  4. 1moOrganic traffic & CTR
    highpull
  5. 1moAI search visibility (AEO / GEO)
    lowpull
  6. 1moApp store ratings & ASO
    medpull
  7. 3moDomain authority / backlink profile
    medpull
  8. 6moContent decay / refresh signal
    medpull

Is demand converting into revenue?

Pipeline is the leading number, cycle length the friction gauge, and win/loss the only loop that tells you why the machine wins or jams.

  1. 1mo◆ Meetings booked / pipeline
    medpull
  2. 1mo◆ Win/loss analysis
    highpull
  3. 3moPipeline stage conversion
    medpull
  4. 3moSales cycle length
    medpull

Team

Can we sustain the people doing it?

Is the hiring machine working?

A ladder with a famously slow truth: response rates answer in days, offers in weeks — but whether the interview signal was right takes two quarters to know.

  1. 3d◆ Sourcing response rate
    medpull
  2. 1wk◆ Interview signal
    lowpull
  3. 1wkOnboarding ramp / time-to-first-commit
    medpull
  4. 1mo◆ Offer acceptance rate
    medpull
  5. 1mo◆ Time-to-hire
    medpull
  6. 6mo◆ Quality of hire / ramp time
    medpull
  7. 6mo◆ Compensation benchmarking
    medpull

Is the team healthy and sustainable?

Morale measured at every cadence — the weekly 1:1, the sprint retro, the quarterly survey — with attrition as the lagging number nothing here should let you reach.

  1. 1wk◆ 1:1 sentiment
    medpull
  2. 1wkPulse surveys
    medpull
  3. 1wkMeeting load / focus time
    medpull
  4. 1wkOn-call load / pages per person
    highpull
  5. 2wk◆ Team retrospectives
    medpull
  6. 1mo◆ Developer experience survey
    medpull
  7. 3moeNPS / engagement surveys
    lowpull
  8. 3moBus factor / knowledge concentration
    medpull
  9. 1yr◆ Attrition / regretted departures
    highpush

Are people growing here?

Nothing in the twice-a-year review should be news if the faster loops are working — its real job is calibration, not discovery.

  1. 3mo◆ Performance reviews / 360 feedback
    medpull
  2. 6mo◆ Growth / promotion readiness
    medpull

Are we working on the right things?

The strategy apex. Every other decision on this page optimises inside a frame; this is the loop that questions the frame.

  1. 2wk◆ Opportunity scoring (importance vs satisfaction)
    medpull
  2. 3mo◆ OKR check-ins / goal scoring
    medpull

Risk & Moat

Are we allowed to keep winning?

Do the unit economics actually work?

Whether the machine makes money per unit: margin as the ceiling, CAC payback as the honest operating metric, LTV as the forecast dressed as a measurement.

  1. 1wkAI / inference spend
    highpush
  2. 3moGross margin
    highpull
  3. 3moBlended CAC
    medpull
  4. 3moCAC payback period
    medpull
  5. 3moPricing realization / discount rate
    highpull
  6. 1yrLTV
    lowpull
  7. 1yrLTV:CAC ratio
    lowpull

How long can we keep playing?

The cash ladder: cost alerts in days, the 13-week forecast weekly, the close monthly, runway as the scoreboard, and the capital market's opinion when you finally have to ask for more.

  1. 1dCloud spend / burn alerts
    highpush
  2. 1wk◆ 13-week cash flow forecast
    highpull
  3. 1mo◆ Budget vs actuals (monthly close)
    highpull
  4. 1moCash collection / DSO
    highpull
  5. 1mo◆ Investor / fundraising feedback
    lowpull
  6. 3moRunway / burn multiple
    highpull

Is revenue healthy — not just growing?

Growth is the number everyone watches; retention and concentration are the numbers that decide whether it means anything.

  1. 3moMRR / ARR growth
    highpull
  2. 3moRevenue churn / GRR
    highpull
  3. 3moRevenue concentration
    highpull
  4. 3mo◆ Revenue forecast accuracy
    medpull
  5. 6moExpansion / referral revenue
    highpull

What could kill us from outside?

The existential pull loops: platforms, vendors, regulators, fraudsters and competitors all move on their own schedule, and none of them will page you.

  1. 1dFraud / abuse / chargeback rate
    medpush
  2. 1mo◆ Platform / API dependency changes
    lowpull
  3. 1mo◆ Competitor launches
    lowpull
  4. 3mo◆ Vendor / provider concentration
    lowpull
  5. 3mo◆ Compliance audits (SOC2 / ISO)
    highpull
  6. 6mo◆ Regulatory & legal signals
    medpull
  7. 1yrChurn by tenure / lock-in strength
    medpull
  8. 1yr◆ Pricing power
    medpull

A loop can serve more than one decision; a decision is only as good as its slowest trusted loop. Dataset v0.18.2 — 199 loops, every one of them reachable from a decision on this page.