Capture small mundane surprises — they reveal systematic…
Capture small, mundane surprises rather than filtering for 'important' ones, because small surprises reveal systematic blind spots that large surprises obscure.
Why This Is a Rule
Large surprises are obvious and get captured naturally — a project fails, a prediction goes wildly wrong, a relationship changes unexpectedly. Everyone notices and processes the big ones. Small surprises — a task taking slightly longer than expected, a colleague responding differently than predicted, a tool behaving in an unexpected way — are individually trivial and get filtered out as noise.
But small surprises, when accumulated, reveal systematic blind spots that large surprises obscure. Three large surprises look like three independent events. Thirty small surprises clustered in the same domain reveal a structural error in your mental model — a consistent pattern of misprediction that no single large surprise would have surfaced.
The filtering instinct ("that wasn't important enough to capture") is exactly the wrong filter for surprise. Importance selects for dramatic, narrative-friendly events. Small surprises select for prediction errors in your mental models — which is the raw material of learning. The mundane surprise about meeting duration, repeated ten times, might reveal that you systematically underestimate transition costs. No individual instance felt "important." The pattern is critical.
When This Fires
- After any moment of mild surprise — "huh, that was unexpected" — even if it seems trivial
- When reviewing your day and something small didn't go as expected
- During conversations when someone says something slightly unexpected
- Any micro-moment of prediction error, regardless of stakes
Common Failure Mode
Filtering for "important" surprises and missing the structural signal. You capture "the product launch failed" (dramatic, important) but not "the standup ran 5 minutes long again" (mundane, trivial). The product launch might be a one-off. The standup overrun, repeated 20 times, reveals a systematic planning error worth fixing.
The Protocol
For one week, lower your capture threshold for surprises to include anything that felt even slightly unexpected. (1) Any time you think "huh" or "that's odd" or "I didn't expect that" — capture it, no matter how trivial. (2) At the end of the week, review the accumulated surprises. (3) Look for clusters: are multiple small surprises pointing at the same blind spot? (4) Clusters reveal systematic model errors that would never have surfaced from big surprises alone.