The Boycott

What this pack runs.

Organized anger arrives as mentions in hours and as lost sales in weeks, and the meeting is held in between. A room that has the mention count and nothing else will decide on it, and it will decide to do something visible. The two questions it never separates are whether the anger is turning into lost sales at all, and for how long, and both can be answered in the first week from data the company already holds: who is angry and what they buy, from the loyalty data and the order book rather than the comments, and what has actually stopped being bought, from the scan data against the base and the category. This pack turns that into arithmetic. What the angry buy, times the share of it that has actually stopped, at the margin it carried, for as long as the loss stays lost, is what holding the position costs. Against it: the write-off once, what the position earns while it stands, and the second anger, which is what the people who wanted the position held stop buying when it is dropped, and which is frequently the larger half and never in the paper. The decision reduces to the break-even bite, how far the bite can travel before reversing is cheaper than holding, and the break-even absence, how long the loss can stay lost before reversing wins. It stops on two conditions: no one has measured what has stopped being bought from sales rather than mentions, and no one has costed the reversal.

The casebooks these cases come from.

The pack reads nine sourced decisions through the framework. They are drawn from these casebooks, each one a set of verified records on a single decision type.

A verified record has passed a mechanical claim-versus-source check, a second model's read, and human approval. The count is what exists today, not a target.