The account from the post, and the same estimator behind it. Move the
window and watch the effect and its range move with it. The effect is the average gap
between what happened and the projected line, recomputed here as you drag.
The post reads 5, because that is where the projection becomes
vaguer than the metric's own daily swing. Push it out and watch the range widen, and the
estimate drift with it, as the projection climbs away from the last real day. You buy length
with confidence, not with certainty.
Cost per lead, observedWhat it would have done anywaythe change