The Customer Success Signal
Writing on churn prediction, retention ops, and CS automation.
Usage Drop vs. Support Spike: Which Churn Signal Matters More?
Most CS teams pick one churn signal to track. We analyzed both across 300 accounts to find out which one predicts non-renewal more reliably — and when neither alone is enough.
How to Build CSM Playbooks That Actually Run Without You
A playbook that depends on a CSM remembering to look is not a playbook — it's a to-do list. Here's how to design automated trigger logic that fires reliably at scale.
Mid-Market B2B SaaS Retention Benchmarks: What Good Actually Looks Like
Net revenue retention above 100% gets all the press. But for CS teams managing 200–1000 accounts, what does a realistic retention floor look like — and how do you know if you're below it?
PLG Companies Have a Churn Blind Spot in Their Account Tier
Product-led growth companies watch individual user activation obsessively. But when a paying account has 3 users and 2 go quiet, the account-level signal often goes untracked until renewal.
Your Health Score Is Lying When Your Support Queue Is Spiking
A CRM health score built only on product usage looks green while the customer is writing their third escalated support ticket.
A Practical Guide to Renewal Forecast Accuracy for CS Teams
Renewal forecasting that's wrong 40% of the time is barely better than guessing. This guide covers the data inputs and scoring approaches that measurably improve forecast accuracy.
CSM Automation Is Not CSM Replacement — Here's the Difference
Every CS automation tool gets the same question from team leads: 'Is this going to replace my CSMs?' The honest answer — and why automation that reduces manual triage actually makes CSMs more effective.
The Signals That Predict Expansion Revenue Before Your CSM Asks
The same usage telemetry that flags at-risk accounts can also identify expansion-ready accounts — often 30–45 days before the CSM's quarterly check-in.
NPS Is a Brand Signal, Not a Churn Predictor
If your churn early-warning system is built on NPS scores, you're reading the wrong gauge. NPS measures satisfaction at a point in time. Churn behavior unfolds over weeks of behavioral signals that NPS never captures.
What Pendo Tells You About Churn — And What It Misses
Pendo is excellent for in-app analytics and user guidance. But CS teams using Pendo as their sole churn signal source are missing the support-side picture.
How to Measure the Accuracy of Your Churn Prediction Model
Precision vs. recall in churn prediction isn't just an ML concept — it has direct business implications. High false positives waste CSM time on healthy accounts.
The CS Ops Stack in 2026: What You Actually Need vs. What Gets Sold
CS platform vendors want to be your entire operating system. But most mid-market CS teams run better on a focused stack. A guide to not over-building.
When to Reach Out: The Timing Science Behind Proactive CSM Intervention
Reaching out too early looks presumptuous. Too late and the account's already decided to leave. Data from CS teams using behavioral triggers shows a 3–5 week window is the intervention sweet spot.