First of the Month: Database Forensics

Most operators open the May dashboard on the first and start thinking about May.

The contrarian claim is this. The April book is more useful than the May pipeline on day one. The pipeline feels urgent because it’s what’s coming. The closed book feels finished because the month is over. The finished book has more decision-relevant information than the pipeline does, and almost nobody reads it.

The book you closed last month isn’t a settled number. It’s a forecast of the next ninety days, and most of the forecast is already legible if you sit down with the list and read it. Three columns do most of the work.

The first is lead origination date. For every member who signed in April, when did the lead first hit the CRM? A close that came from a lead aged thirty days is a different member than a close that came from a lead aged three days. The thirty-day lead thought about it, came back, signed deliberately. The three-day lead is closer to an impulse close, which is fine, but impulse closes have higher early-cancel risk. If April’s book is heavy on short-aged closes, May’s cancellations are going to skew toward the front of the cohort. You can see that on day one. You can’t see it on day forty-five when the cancels are already in.

The second is promo versus baseline. Sort April’s closes by what they signed under. Promo closes and baseline closes don’t behave the same way over ninety days. The promo close came in on a price-driven decision, which means the price is doing more of the work in keeping him there. When the promo period ends and the rate steps up, a portion of the cohort leaves. That’s not a surprise, it’s a structural feature of how promos work. What matters on day one is the ratio. If April was sixty percent promo, the May 1 conversation is about which of those members are likely to step up cleanly and which ones aren’t. If April was ninety percent promo, you have a different problem, and it’s not a May problem, it’s an acquisition problem that’s going to surface in July.

The third is the soft close. Not every signed agreement is a real close. Some of them are reps closing prospects who weren’t ready, getting the signature, and producing a member who’s going to cancel inside thirty days. The soft close looks the same as a real close on the dashboard. It looks different in the database if you know what to look for. Short tour, no second visit, signed at the end of a long conversation, paid the minimum. None of those alone mean anything. Two or three of them on the same agreement is a signal. Pull the soft closes out of April’s book and you have a reasonable list of who’s most likely to cancel in May.

This is what day-one forensics is. Not a dashboard review. A read of the actual book, with the right columns visible, looking for the signal that tells you which part of April is going to come back as a churn line in May.

The work is maybe an hour if the CRM is clean. Two if it isn’t. Most operators don’t do it, because the pipeline is louder and the new month is more interesting than the old one. The cost of skipping it is that May’s cancellations arrive as a surprise instead of as something you saw coming on the first.

The first is the right day for this. The numbers are settled, the noise is gone, the month hasn’t started running yet. The pipeline will still be there at noon. Read April first.