Complexity Determines Convergence Time, Not Depth
Complexity isn’t about doing MORE. It’s about how long it takes for research to CONVERGE — for answers to stop changing your tentative decision.
The Reframe
| Old Framing | New Framing |
|---|---|
| ”Complexity means do more things" | "Complexity means budget more time for the same process to converge” |
| Creates obligation | Gives permission to stop |
| Framework becomes burden | Framework enables closure |
The actions are the same in each quadrant. The time to “done” varies.
What Complexity Actually Affects
| Quadrant | What You DO | What Complexity Affects |
|---|---|---|
| Q1: HEURISTIC | Gut check → decide | How quickly you identify “good enough” |
| Q2: SYSTEMATIZE | Build a rule/system | How many iterations before edge cases are handled |
| Q3: CAREFUL | Questions → research → decide | How many “that answer raised a new question” loops |
| Q4: BUILD CAPABILITY | Build knowledge structure | How sprawling the domain is (scope of KB) |
Complexity Budgets
| Quadrant | Low Complexity | High Complexity |
|---|---|---|
| Q1 | 30 sec – 2 min | 2 – 5 min |
| Q2 | 15 min design, 1 week test | 30 min design, 2 week test, 1 revision cycle |
| Q3 | 3-question coverage, 15 min/question | 5-question coverage, 45 min/question |
| Q4 | 5-10 item reference sheet | Scoped KB (define boundaries first) |
The Budget Is Permission to Stop
The framework can push you into infinite research. Knowing about biases makes you feel like you should check for biases. Knowing about Seven Lenses makes you feel like you should run all seven.
The complexity modifier doesn’t say “do more.” It says “you’re ALLOWED to stop after this much.”
| Complexity | What the Budget Means |
|---|---|
| Low | ”You’re allowed to stop after 30 min of Q3” |
| High | ”You’re allowed to stop after 2 days of Q3” |
In both cases: permission to STOP, not obligation to CONTINUE.
Finishing early is fine. The budget gives you permission to stop, not a target to fill.
How to Assess Complexity
Use Objective Complexity Signals — external signals, not feelings:
| Signal | Low (0 pts) | High (1 pt) |
|---|---|---|
| Precedent | Done this type before | Novel territory |
| Option clarity | 2-3 obvious choices | Many unclear options |
| Variable familiarity | Understand interactions | Can’t model interdependencies |
| Expert consensus | Experts agree | Experts disagree |
| Reversibility | Easy to undo | Locks in downstream |
| Stakeholders | Just me / aligned | Multiple conflicting interests |
Score: 0-2 points = Low complexity budget. 3+ points = High complexity budget.
Common Trap
Using complexity to avoid Q1 decisions. “This is complex” can become an excuse to over-analyze low-stakes choices. The question isn’t “is this complex?” — it’s “do the stakes justify engaging with the complexity?”
Low stakes + high complexity = still Q1 or Q2. You satisfice on the complexity rather than resolving it.
North: Where this comes from
- Stakes × Recurrence Quadrant Model (the framework this modifies)
- The Weighing Problem (why budgets matter)
East: What opposes this?
- Complexity as Depth (“more complex = do more things”)
- Perfectionist Research (no stopping point)
South: Where this leads
- Coverage Then Clock Structure (how to structure time within budget)
- Counter-Weight Rules by Default Type (how budget applies differently by default)
West: What’s similar?
- Time-Boxing in Agile (constrained iteration periods)
- Satisficing (good enough vs. optimal)
- Parkinson’s Law (work expands to fill time available)