Customer health has always been a cornerstone of Customer Success — yet the way most organisations measure and act on it hasn’t evolved fast enough.

Over the past few months, I’ve explored this transformation — from static health scores to dynamic, AI-driven systems that connect data to outcomes. This post brings all the pieces together in one place and looks ahead to what comes next.

1️⃣ Why Customer Health Is Broken

Most health scores are static and misleading. A “green” account can still churn because the score doesn’t capture context or timing.
👉 Read the post →

Customer Success teams don’t struggle with a lack of data — they struggle with knowing what to do next.

Dashboards are full of signals, metrics, and insights, but too often they sit unused. The problem isn’t visibility — it’s action.

That’s why well-designed alerts are such a powerful bridge. They translate data into movement, ensuring that when something changes in the customer relationship, the right person knows, at the right time, and does the right thing.

Over the last few posts, we’ve explored how customer health has evolved — from static scores to dynamic systems, and now to AI-powered prediction:

• In Why Customer Health Is Broken (and How to Fix It), I discussed why traditional health scores fail to capture real customer dynamics.
• In From Health Score to Health System, we looked at how to turn scores into signal-driven systems.
• And in AI Customer Health Scoring: How to Build It (and How to Avoid Bias), I explained how AI adds predictive intelligence to those systems.

Now, let’s close the loop — because even the best health models fail if alerts don’t trigger the right action.

Customer health has always been at the heart of Customer Success. But in most SaaS organisations, the way we measure it hasn’t kept up with how customers actually evolve.

We started with health scores — simple, single-number snapshots that promised clarity but often delivered false confidence. Then we learned that what really matters isn’t the score itself but the system behind it — a structure of signals, alerts, and playbooks that help teams act in real time.

That evolution took us from monitoring customers to actually managing them. Still, even the best health systems face a new challenge: scale.

When you’re tracking thousands of customers across dozens of products, regions, and behaviours, no human can realistically spot every trend or risk early enough. That’s where AI health scoring comes in — not as a replacement for human judgment, but as an accelerator of it.

AI gives us the ability to move from static, manual interpretation to predictive, data-driven action. Instead of waiting for a red score or a missed milestone, we can now see subtle behavioural shifts — declining engagement, tone changes in communication, or anomalies in product usage — that reveal what’s coming next.

Most SaaS companies track customer health using a simple score — green, yellow, red.

It feels neat and manageable. But here’s the reality:

• A customer can be “green” today and churn tomorrow.
• Scores often lag behind reality.
• They rarely capture the complexity of relationships, adoption, and business context.

Health isn’t static. It’s dynamic, evolving every day. And health scores have limits. 

That’s why we need to move beyond a health score into a health system.

What’s Wrong with Health Scores?

If you're in Customer Success, you've likely seen this before: A customer with a “green” health score churns without warning. Or worse — your team scrambles to rescue an account, but the health score never reflected the risk.

Why? Because traditional health models are broken.

They’re often static, backward-looking, and overly simplified. A single score can’t capture the complexity of real customer relationships, especially in high-growth SaaS environments where product usage, stakeholder dynamics, and business shifts change rapidly. I've observed this at HubSpot, Cisco and now at ServiceNow.

With the advent of AI, it’s time for a better way.

The Problem with Traditional Health Scores

Most health scores today rely on a formula of 3–5 inputs: