book-triagelisted

# Book Triage Even coverage is not the goal. It is the failure. A book of forty accounts touched equally is forty accounts touched too lightly to change anything, and the CSM doing it feels busy and is not moving a single outcome. The job is to decide where attention goes and, harder, to write down where it does not. The failure this exists to prevent: **the week spent on the accounts that asked.** Attention allocated by who emailed loudest, which correlates with neither risk nor value, and which systematically starves the quiet accounts where the actual danger is. --- ## What this needs **Shared context.** If an `account-context` document exists, read it first: the segments, the contract shapes and what healthy usage looks like here. Without it, "usage dropped" cannot be told apart from a normal seasonal pattern. Where it is absent, carry on and name the assumption. **Minimum: your account list, with renewal dates and rough values.** Enough to triage, though it will lean on what you know rather than on data. **Better with** usage trend, support activity and last-contact dates. The last one is the single most valuable field and the one most often missing. **Best with** health scores you know the composition of, and a stakeholder picture, so movement can be read rather than guessed at. --- ## Step 1: Sort by movement, not by state The most common triage error is ranking by health colour or by size, both of which are largely static. A red account that has been red f