How to Calculate Customer Health Scores for SaaS

Why one churned customer can wreck your month (when you only have 30)

When you're running a small SaaS, every cancellation hits hard. Lose three customers in a month and you've just kissed goodbye to 10% of your base. At $50/month each, that's $1,800 in annual revenue gone—before you even knew they were unhappy.

Here's the kicker: most of those customers gave you signals. Maybe they stopped logging in two weeks ago. Maybe they opened a support ticket asking about export options. Maybe their last email reply was shorter than usual. You just didn't have a system to notice. Without health scoring, at-risk customers blend into the noise until the cancellation email arrives.

The frustrating part? Most churn is preventable. Research shows that companies with formal health scoring see 15–20% higher accuracy identifying at-risk accounts. The problem is that 84% of companies with customer success programs use health scoring—but only 42% of CS teams actually track them. There's a gap between knowing you should do this and actually doing it.

For a solo founder, that gap often looks like: "I'll get to it when I have more customers." But the opposite is true. The smaller your base, the more each customer matters. The sooner you start scoring health, the sooner you can reach out before someone goes silent and cancels. This post gives you a framework you can implement today—no CS platform, no consultants. Just a spreadsheet, three metrics, and 15 minutes.

What a customer health score actually is (in plain English)

A customer health score is a single number that tells you how likely a customer is to stay or leave. High score = they're engaged and likely to renew. Low score = something's wrong and they might churn.

That's it. You're not building a PhD thesis. You're creating a simple early-warning system so you know who needs attention this week. Instead of scanning through your entire customer list every Monday (or worse, not scanning at all), you sort by score and the at-risk accounts rise to the top. One number replaces guesswork.

The traffic-light version vs. the 100-point version

Most systems use one of two approaches:

Traffic-light (Green / Yellow / Red): Quick to eyeball. Healthy customers are green. At-risk are yellow. Critical are red. Perfect when you're reviewing 40 customers on a Monday morning. You can scan a list and know who to call first without doing math.

100-point scale: More granular. Scores from 0–100 let you rank customers and spot trends over time. Common thresholds: Healthy (71–100), At Risk (31–70), Critical (0–30). Enterprise tools like Gainsight and ChurnZero use this, but you can replicate it in a spreadsheet. The number also helps you track improvement—if a customer goes from 45 to 72 over two months, you'll see it.

Both work. Many founders start with traffic lights for the weekly review and add the numeric score when they want to track changes over time. The 100-point scale also makes it easier to compare across customers—who's your healthiest? Who's sliding?—without remembering a bunch of qualitative notes. Start with traffic lights for speed; add a 100-point number if you want more precision. We'll build the 100-point version in the spreadsheet section so you get both.

The solo founder's 3-metric starter framework

Enterprise platforms like Gainsight weight four or more factors—product usage, support trends, sentiment, executive engagement. ChurnZero suggests 5–7 factors, each worth 10–20 points on a 100-point scale. That's overkill when you have fewer than 100 customers. You don't have a CS team to maintain executive engagement scores. You don't have NPS survey infrastructure yet. You need something you can actually maintain—something that takes 15 minutes a week, not 15 hours.

Here's a simpler framework—three metrics, clear weights, no CS team required. It's designed for the founder who's still doing support, sales, and product.

Metric 1 — Product usage frequency (logins/week or core action count)

How often does this customer use your product? For most B2B SaaS, logins per week or a core action count is enough. Your "core action" depends on your product: for a project management tool it might be tasks completed; for an analytics product, reports run; for a design tool, files exported. Pick the one action that best predicts someone is getting value.

Scoring: 5+ logins/week (or equivalent core actions) = 100. 3–4 = 75. 1–2 = 50. 0 in the last 2 weeks = 25. Adjust the thresholds to match your product—if your tool is used weekly, not daily, scale accordingly.

This metric correlates strongly with retention. As our guide to churn warning signs shows, declining login frequency is one of the earliest signals that a customer is about to leave. Product usage is your leading indicator.

Metric 2 — Support signal (positive if none, negative if high)

Support tickets can mean two things: they're engaged and need help, or they're frustrated. For a small base, the signal is usually frustration. One or two tickets? Fine. A spike in tickets, repeated issues, or questions about cancellation? That's a negative signal.

Scoring: 0–1 tickets in last 30 days = 100. 2–3 = 75. 4+ or any cancellation-related inquiry = 50. Multiple escalations = 25. If a customer asks "how do I export my data" or "what happens if I cancel," that's a yellow flag even if they only have one ticket.

Process Street, a real SaaS company, tracked just two metrics (checklist runs + web sessions) because Gainsight cost $40,000/year. As one of their team members put it: "We couldn't justify $40K for a platform when we could get 80% of the value from a simple dashboard." They proved you don't need a complex system to improve retention.

Metric 3 — Your gut-check score (1–5 based on conversations)

You talk to your customers. You know who's excited and who's likely to churn. Turn that intuition into a number.

Scoring: After each significant conversation (email, call, demo), rate them 1–5. 5 = "Would definitely renew." 1 = "I'm worried they're leaving." Scale to 100 by multiplying by 20. Update gut-check within 24 hours of the conversation—your memory fades fast. If you haven't talked to a customer in 60+ days, leave the score blank or use your last rating; don't guess.

This metric captures what spreadsheets can't: tone, context, relationship strength. It's less scientific but often the most predictive. When your gut says something's off, it usually is. (And if you're also dealing with failed payments, our dunning best practices can help you recover revenue while you fix the relationship.)

How to weight them (50/30/20 split)

Product usage: 50% — Most predictive of retention. Easy to track. Start here.

Support signal: 30% — Catches frustrated customers before they cancel. Worth more when support load is high.

Gut-check: 20% — Your secret weapon. You're the only one who can score this. Use it.

Weighted example: Usage 75 × 0.5 + Support 100 × 0.3 + Gut 80 × 0.2 = 37.5 + 30 + 16 = 83.5 → Healthy.

Building the actual spreadsheet in 15 minutes

You can build this in Google Sheets in about 15 minutes. Here's the step-by-step.

Step 1: Set up your columns

Step 2: Add the weighted formula

In column G, use: =D2*0.5+E2*0.3+F2*0.2

Drag down for all rows.

Step 3: Add conditional formatting for status

In column H, use: =IF(G2>=71,"Healthy",IF(G2>=31,"At Risk","Critical"))

Step 4: Color-code the score column

Apply conditional formatting to column G:

• Green: 71–100
• Yellow: 31–70
• Red: 0–30

Step 5: Populate with real data

Pull usage from your analytics (logins, core actions). Pull support data from your help desk. Update gut-check after each customer conversation. Refresh weekly.

Here's what a few rows might look like for a fictional 40-customer SaaS selling at $50/month:

Pro tip: Add a "Last updated" cell so you know when you last refreshed. Stale data defeats the purpose. Set a recurring Monday morning reminder to update the sheet—15 minutes a week is all it takes. If you find yourself skipping weeks, simplify further: drop gut-check temporarily and just track usage + support until the habit sticks.

Weekly workflow: Every Monday, update usage and support scores from your tools, refresh gut-check for anyone you talked to last week, and sort by score ascending. Your critical and at-risk customers rise to the top. Pick the top 3–5 and reach out that day. A quick "Hey, noticed you haven't logged in lately—everything going okay?" can save a renewal. The goal isn't to fix everyone at once; it's to never let a customer slip through unnoticed for three weeks.

When to level up from 3 metrics to 5

Stay with three metrics until one of these triggers:

• You pass 50 customers — Manual tracking gets noisy. Add structure.
• You add a second product line — Usage across products matters.
• Gut-checks feel unreliable — Maybe you're not talking to everyone enough.

When you level up, add:

• NPS (Net Promoter Score) — One question: "How likely are you to recommend us?" 0–6 = detractors, 7–8 = passive, 9–10 = promoters. Survey quarterly. Simple to add, hard to game.
• Feature adoption depth — Are they using your best features or just the basics? A customer who only uses the free-tier features on a paid plan is at higher risk. Low adoption = higher churn risk.
• Onboarding completion — Did they finish setup? Did they connect their integrations? Incomplete onboarding correlates strongly with early churn. Track this from day one—customers who don't activate in the first 30 days often never do.

Gainsight's weighted model uses Product Usage (40%), Support Trends (25%), Sentiment (20%), and Executive Engagement (15%). You can emulate that later—but start simple.

The numbers behind health scoring

Health scoring isn't theoretical. The data backs it up.

• 5% retention improvement → 25–95% profit increase (Bain & Company) — Bain's research showed that a 5% improvement in customer retention can increase profits by 25% to 95%, depending on the industry. For SaaS, the multiplier is on the higher end. Small gains in retention have outsized impact on profitability.
• 15–20% higher accuracy identifying at-risk accounts for companies with formal health scoring (Totango) — Totango's research found that teams using structured health scores catch at-risk customers earlier and more consistently. You stop relying on "who do I remember being quiet lately" and start relying on data.
• 35% churn reduction for SmartReach after implementing weighted health scoring — SmartReach, an outreach platform, reported a 35% drop in churn after rolling out a weighted health scoring model. They identified at-risk accounts faster and intervened before cancellations.
• 84% vs. 42% — 84% of companies with customer success programs say they use health scoring, but only 42% of CS teams actually track them consistently. That gap is your opportunity. Most of your competitors aren't doing this well—and the ones who are often overcomplicate it with enterprise tools you don't need yet.

Avoid these mistakes: Don't over-weight gut-check (it's subjective). Don't skip the weekly refresh (stale scores are worse than no scores). Don't chase every yellow customer—focus on critical first, then at-risk. And don't build a 10-metric monster. Start with three. Add more only when you outgrow them.

---

Health scoring doesn't require a CS team or enterprise software. Start with three metrics, build your spreadsheet in 15 minutes, and start catching at-risk customers before they churn. The companies that do this well see better retention. The ones that don't often wonder why their churn spiked. You don't have to be in the second group.

When you're ready to automate—when the spreadsheet takes too long or you want alerts the moment a score drops—tools like Tether can sync your usage data, calculate health scores automatically, and notify you in Slack. But until then, a spreadsheet and 15 minutes a week will get you most of the way there. The goal isn't to have the perfect system. It's to have a system you'll actually use. Start simple. Add complexity only when you need it.