Tracking triggers
Most trigger tracking fails because it records the trigger and nothing else. The event is usually the least informative part — the same message produces a three on Tuesday and a nine on Friday, and what differs is everything that was already there.
What to record, and in what order
Seven fields. The first four are the ones almost everyone leaves out, and they are the ones that explain the variance.
- Hours slept last night, and whether you have eaten today
- What was already unresolved this week — a conversation, a deadline, a silence
- Where you were on a scale of one to ten before anything happened
- Time of day
- The event itself, stated as a camera would record it, with no interpretation
- The intensity it produced, out of ten, and how long until it halved
- What you did, and whether anything helped
What a month of this usually shows
Five findings that appear in most people's logs, and that are close to invisible without writing.
| What shows up | What to do with it |
|---|---|
| The pre-load explains more than the event | Same trigger, wildly different outcomes, sorted almost entirely by sleep and by what was already unresolved. This moves the work upstream, to things you can actually control. |
| Two or three triggers account for most of it | Usually ambiguity in one relationship, one recurring situation, and one time of day. Naming them specifically is more useful than a long list. |
| A time-of-day pattern | Late evening for most people. Not mysterious — capacity is lowest and ambiguity is highest when nobody is replying. It is also the most schedulable thing you will find. |
| The intensity you start the day at is predictive | Logging a morning number takes ten seconds and predicts the day better than anything else you can record. It also tells you which days to be careful with. |
| The skills that worked are not the ones you expected | Written down, the things that actually helped are often unglamorous — leaving, eating, a walk. The named skills tend to underperform their reputation, and that is worth knowing. |
Making it work
The practical problems with tracking anything for a month.
Doesn't focusing on triggers make it worse?
Recording is different from ruminating. A log is a fixed set of fields completed in ninety seconds; rumination is open-ended and does not end. If the log starts becoming an essay, shorten it.
What if I can't identify the trigger?
Write 'nothing identifiable'. That is a real and common entry, and a cluster of them is itself a finding — often pointing at the pre-load rather than at anything external.
How long before it's useful?
Two weeks for the first patterns, a month for confidence. Fewer than ten entries will not show you anything, which is why most attempts conclude nothing before they had a chance.
Should I show this to anyone?
A clinician, yes — dated entries change what an appointment can do. It is also the single best answer to being dismissed, because a written record is much harder to talk past than a description.
What if the trigger is one person?
Then that is the finding, and it is worth taking seriously rather than continuing to track. Whether that is a favorite-person dynamic, an unreasonable situation, or both, the log has done its job.
Where does the book cover this?
Chapter 9 on body awareness is the noticing half; chapter 21 on sustaining recovery is the pattern-spotting half. Neither prescribes a log, and both depend on the information one produces.