When a CS team discovers a renewal is at risk less than 30 days before the renewal date, the save rate drops below 20%. Most teams find out even later, when procurement stalls or the champion stops responding. By then, the decision has already been made. The gap between an at-risk signal firing and a save play landing on someone's queue is the single largest preventable source of churn in most revenue orgs.
Most renewal-risk identification is reactive. A CSM notices the customer has not responded to emails. A leader sees a renewal on the board that has not been discussed. Someone remembers a support escalation from three months ago. By the time these observations converge into a "this account is at risk" conversation, the customer has already started evaluating alternatives or decided to let the contract lapse.
The underlying problem is structural, not individual. Without a scoring model that continuously weighs multiple risk variables (usage trends, stakeholder health, support burden, contracting momentum, executive engagement, and market conditions), the team is left relying on memory and instinct. Instinct does not scale past 50 accounts per CSM.
A working renewal risk model is a weighted composite. Eight inputs, tuned per segment, computed daily. Each input fires as its own signal, and the composite triggers the save-play routing.
The composite fires against severity thresholds. Above the critical threshold, the account moves to T2 At-Risk and a save play is assigned. Below it, the account stays in T3 Monitor and compounds are watched.
Renewal Risk is one of PILLAR's five scoring formulas. Every renewal is scored daily. The renewal_save play is a named template in the library with measured win rates. The triage board visualizes the full portfolio state. Saves feed back into scoring weights on a governed cadence.
Your Blueprint showed your Renewal & Retention score. Want to understand why it scored the way it did - and what to do about it in the next 30 days?
Get Your Free Blueprint
The benchmark figures here are PILLAR operator estimates. They come from our own work with EdTech and public sector revenue teams, not from a published study and not from a survey. They are a practitioner's calibration, offered so the argument has something concrete to push against, and they should be read that way rather than cited as measurements.
Where a figure on this site does come from a public record, it is sourced at the point it appears. Anything drawn from a state disclosure file, a federal dataset or an independent outcome measure carries its source, its vintage and a way to re-derive it.
Every figure PILLAR publishes, on this page and everywhere else, is covered by our standing correction offer. If one of these estimates does not match what you see in your own book, we want to hear it.