Customer Health Score Calculator
Score your CS book across 5 weighted dimensions. See at-risk ARR, get a portfolio grade, drill into Red accounts, and triage your Monday-morning save motion. Free, no signup.
Last reviewed: April 2026
Weighting (auto-normalizes to 100)
Σ = 100Customers (5 / 100)
| Customer | MRR | Seg | U | E | B | S | Sp | Comp | |
|---|---|---|---|---|---|---|---|---|---|
| 79 | |||||||||
| 53 | |||||||||
| 34 | |||||||||
| 68 | |||||||||
| 90 |
Portfolio Heatmap
Portfolio Report Card
ARR Distribution
Intervention Queue(ARR × inverse-score — lose this first)
What-If: Shift Weights
Support = remainder to 100%. Live preview below — click Apply to commit.
Reverse Calculator
Scenario A vs B Compare
Session History
What is a customer health score, and why does weighting matter?
A customer health score is a single 0–100 number that compresses everything you know about a customer — product usage, sentiment, billing behaviour, champion stability, support load — into an at-a-glance indicator of churn risk. It's the most important metric in customer success, and almost every CSM spreadsheet in the world is some version of it.
Weighting is where most health scores fail. If you weight all five dimensions equally, you're claiming that a billing delay predicts churn as strongly as a product-usage collapse — which is false for most SaaS businesses. The right weights depend on your motion. SMB SaaS weights Usage and Engagement heavily (60–70% combined); Enterprise weights Stakeholder 30–40% because champion loss drives churn more than usage dips; agencies weight Billing because service-heavy contracts churn on payment disputes.
This customer health score calculator for SaaS lets you tune all five dimension weights live and watch the portfolio reshuffle in real time — the fastest way to calibrate a health score template without writing a line of SQL.
The 5 dimensions of a weighted customer health score
A multi-factor customer score bundles signals from different systems into one composite. The five dimensions used here are the industry standard taught by Gainsight Pulse, SuccessCOACHING, and CSM Practice.
Usage (default 30%)
Logins per week, core-feature adoption, depth-of-use. Pulled from product analytics (Mixpanel, Amplitude, June, Heap). Strongest predictor for PLG and SMB SaaS.
Engagement (default 25%)
NPS, CSAT, sentiment from QBR notes, attendance at onboarding calls. Pulled from Delighted, Wootric, or your CSM notes. Lagging indicator — great for confirming trends, weak for catching early drops.
Billing (default 20%)
On-time payment rate, MRR trajectory (expansion vs contraction), auto-renew status, days-to-pay. Pulled from Stripe / Chargebee. Blunt but very reliable — billing delays almost always precede churn.
Stakeholder (default 15%)
Champion strength, executive sponsor engagement, org-chart stability. Pulled from LinkedIn monitoring (UserGems, Champify), CRM. Dominant predictor for Enterprise.
Support (default 10%)
Ticket volume, severity distribution, time-to-resolution, repeated-issue rate. Pulled from Zendesk / Intercom. Non-linear — both very high and zero support volume can predict churn.
The composite formula is straightforward: composite = (Usage × wU + Engagement × wE + Billing × wB + Stakeholder × wS + Support × wP) / 100. Weights auto-normalize to sum to 100 as you adjust them.
How to set dimension weights for SMB, Mid-Market, and Enterprise SaaS
CSM health score weighting is one of the highest-leverage decisions in customer success — and one of the most commonly fumbled. These are the starting points for the three main SaaS motions:
Use the preset chips in the calculator to jump between these weighting schemes and watch how the same customer portfolio reshuffles bands. It's the clearest demonstration of how much weighting tilts verdicts — the same customer can land Green or Amber depending on which motion you score against.
Portfolio health score dashboard: how to read the heatmap
A portfolio health score dashboard lives or dies by how quickly a CSM can triage Monday morning. The heatmap in this tool shows every customer as a colored cell: Green (80–100), Yellow (60–79), Amber (40–59), Red (<40). Click any cell to drill down into per-dimension scores, recommended playbook, and a copy-paste outreach email.
The heatmap is sortable by composite, MRR, or band — and once you sort by MRR, a brutal truth appears: one Red $500K account matters more than ten Red $5K accounts. The intervention queue below the heatmap ranks customers by MRR × inverse-composite (dollars × bad-ness) — the "lose-this-first" queue every CS leader needs but almost nobody computes.
Tracking at-risk ARR: the only metric that matters
At-risk ARR is the sum of annual MRR across all Amber and Red customers. It's the single most important number on a CSM's dashboard because it answers the question boards actually ask: "how much ARR could we lose this quarter if nothing changes?"
The at-risk ARR calculator for CSMs in this tool applies a zone override: if red-band ARR exceeds 15% of total portfolio ARR, the zone is forced to Critical regardless of composite grade. The dollars matter more than the count. A VP of CS with 5% Red accounts but 30% Red ARR is in a dramatically worse position than one with 20% Red accounts but 5% Red ARR.
Track at-risk ARR weekly, not monthly. The difference between a 14-day save motion and a 45-day save motion is often the difference between retention and churn. This tool's session history stores up to 10 weekly snapshots so you can see the trajectory — the sparkline in the history panel is calibrated for exactly this use case.
Customer health score formula (the simple weighted average)
The customer health score formula used here is the closed-form weighted average:
composite = (uScore·wU + eScore·wE + bScore·wB + sScore·wS + pScore·wP) / 100
where each score is 0–100 and weights (wU + wE + wB + wS + wP) = 100.
Portfolio composite is ARR-weighted, not unweighted: portfolio = Σ(composite × MRR × 12) / Σ(MRR × 12). This protects against a deceptive "high average, bad portfolio" pattern where many small healthy accounts mask a few huge unhealthy ones.
Band thresholds are 80 / 60 / 40 — round numbers that align with most industry rubrics (Gainsight, Totango, ChurnZero use near-identical cutoffs). Feel free to adjust mentally for your own risk tolerance, but round numbers make board decks easier to read.
Customer success score builder vs Gainsight: when do you graduate?
A free customer success score builder like this covers the core use case — weighted scoring, portfolio view, intervention triage — without the $50K-a-year license. Most CS teams under $5M ARR can run a health score in a spreadsheet or a tool like this and never touch Gainsight.
You graduate to a real CS platform (Gainsight, Vitally, Custify, Planhat, Catalyst, ChurnZero, Totango) when three things become true: (1) your book is > 100 accounts, so manual scoring is untenable; (2) you need automated data pulls from CRM/Stripe/Mixpanel because weekly data entry is a full-time job; (3) you need workflow automation — Red-band customers auto-escalated, auto-assigned playbooks, auto-generated QBR decks.
Until then, a customer health score template + a weekly Monday triage session + this calculator is more than enough. In fact, building your scoring rubric by hand before adopting a platform means you'll know which weights and dimensions actually predict churn — most platforms make it too easy to accept the default rubric without questioning it.
Building a customer health score spreadsheet (and when to stop)
Every CS team starts with a customer health score spreadsheet. The first version is usually a Google Sheet with customer name, MRR, and a traffic-light column. The second version adds five dimension columns, a weighted-sum formula, and conditional formatting. The third version tries to pull data from Stripe and Mixpanel via a rickety Zapier workflow.
Spreadsheets hit three breaking points: (1) data entry becomes a bottleneck above ~40 customers; (2) version control across CSMs becomes chaos (whose spreadsheet is authoritative?); (3) trend analysis requires snapshotting the spreadsheet weekly, which nobody remembers to do. This calculator solves all three by running in the browser, persisting to localStorage, and keeping a 10-session history automatically.
When the browser + localStorage approach breaks — when you need multi-CSM shared state, automated data ingestion, or ML-driven churn prediction — that's when you buy a CS platform. Not before.
How to weight customer health dimensions for your motion
The quickest way to weight customer health dimensions correctly is to backtest against your actual churn. Pull your last 12 months of churned accounts, score each on the 5 dimensions using the data you had 90 days before they churned, and find the weight vector that maximizes the correlation between composite and churn outcome.
If you don't have that data yet, the Reverse Calc mode in this tool supports a simpler calibration: enter "previous composite" for a handful of customers and the tool will grid-search the weight space to find the weight vector that best predicts your prior scores. It's not a churn model, but it's a decent proxy for "what weighting is my team actually using in practice?"
Recalibrate weights quarterly, not more often. Health score weights should be stable enough that customers can be compared across quarters. If your weights shift dramatically every month, your scoring rubric is a vibe check, not a metric.
Frequently Asked Questions
How do you calculate a customer health score in SaaS?
A customer health score is a weighted 0–100 composite across multiple dimensions. Formula: composite = (usage·wU + engagement·wE + billing·wB + stakeholder·wS + support·wP) / 100, where each dimension is scored 0–100 and weights sum to 100. Customer with scores 85/70/90/60/75 and weights 30/25/20/15/10 composite = 76.
What is a weighted churn risk score?
A 0–100 score combining multiple leading indicators where lower scores predict higher churn probability. Customers below 40 are high-risk; below 60 warrant a CSM intervention. Weights reflect which signals predict churn in your motion — usage for PLG, stakeholder for Enterprise.
What are good CSM health score weights?
Defaults for balanced B2B SaaS: Usage 30%, Engagement 25%, Billing 20%, Stakeholder 15%, Support 10%. SMB/PLG weights Usage + Engagement 55–75%. Enterprise weights Stakeholder 30–40%. Agency/services weights Billing + Stakeholder 50%+. Recalibrate quarterly.
Is there a free Gainsight alternative?
Yes — this calculator covers the core health-scoring use case (weighted scoring, heatmap, at-risk ARR, triage) at zero cost. You lose automated data ingestion, workflow automation, and ML prediction. For books under 100 accounts or teams prototyping before buying a platform, this is sufficient.
How do you predict churn with a customer health score?
Track composite weekly, watch trend deltas (a 10+ point drop in 30 days is an early-warning signal), and escalate any customer below 40. Action must happen 60–90 days before renewal to change the outcome — the health score is a leading indicator, not a post-mortem.
What is a customer health score template?
A template defines the dimensions, scoring rubric, and weights. The 5-dimension template (Usage, Engagement, Billing, Stakeholder, Support) is industry standard, taught by Gainsight Pulse and CSM Practice. Each dimension is scored 0–100 using sub-signals mapped to your data sources.
How do you build a multi-factor customer score?
Start with 3–5 dimensions, not 10. Each additional dimension adds noise faster than signal. Score each on 0–100, normalize scales, weight based on what predicts churn in your data — not wishes. Recalibrate quarterly against actual renewal/churn outcomes.
How do you track at-risk ARR for a portfolio?
At-risk ARR = sum of annual MRR for all Amber + Red customers. Track weekly; pair with the intervention queue (sorted by MRR × inverse-composite) to triage highest-value saves first. One Red $500K account matters more than ten Red $5K accounts — dollars beat counts.