Most creators watch "total members" or "new signups" as their north star. That's a lagging indicator — by the time it drops, you've already lost the revenue.

The number that actually predicts a revenue drop 2-4 weeks out: refund + dispute rate on your highest-priced plan, segmented by cohort.

Here's why: whales don't usually churn quietly. Before a high-LTV member cancels, they often:

• Open a support ticket or dispute first

• Downgrade to a cheaper plan

• Go quiet in your community/chat for 1-2 weeks

If you're only looking at aggregate MRR, none of this shows up until the charge actually fails to renew. By then it's too late to save them.

What to do instead:

1. Segment your refund/dispute data by plan tier, not just totals — a 2% refund rate on your $29 plan means something very different than 2% on your $299 plan.

2. Set an alert threshold for any refund or dispute on your top-tier plan — treat every single one as a "why did this happen" investigation, not a rounding error.

3. Track cohort retention (signup month vs. still-active), not just point-in-time member counts. A cohort curve tells you if a pricing or onboarding change actually worked.

Built WhopPulse specifically because I couldn't find a way to see this across multiple products without exporting CSVs and building my own spreadsheet every week. Happy to answer questions on how creators running multiple products/tiers are tackling this.