If you’re running a B2B Customer Success or Account Management team, you know the pain: figuring out which customers are happy, who’s slipping, and—most importantly—who’s about to churn. Automated health scoring promises to solve this, but most guides are vague or just pitch you features. Here’s a no-nonsense walkthrough for setting up customer health scoring in Vitally that actually works for real B2B teams.
This guide is for CS leaders, ops folks, or anyone who wants to stop guessing and start acting on real customer data. No fluff, no magic formulas—just a practical system you can build, tweak, and actually trust.
Why Bother With Automated Health Scores?
Manually guessing which clients are healthy is a losing game once you have more than a handful. You need a way for your whole team to see—at a glance—who needs attention and who’s on track.
Automated health scoring in Vitally can:
- Give you an early warning when a customer starts to slip
- Help prioritize outreach and renewals
- Let you measure the impact of your team’s work
But—and this is important—health scores are only as smart as the data and logic you put in. Don’t expect magic insights just by flipping a switch.
Step 1: Nail Down What “Health” Actually Means for Your Customers
Before you touch a single setting in Vitally, get clear on what “healthy” looks like for your customers. Skip this, and your scores will be just as random as gut feelings.
Ask yourself:
- What behaviors signal a happy, successful customer?
- Regular logins? Usage of key features? Support tickets (or lack thereof)?
- What’s a red flag?
- Drop in usage, overdue invoices, negative survey feedback, stalled onboarding, etc.
- Are some signals more important than others?
- For example, a big drop in usage probably matters more than a missed newsletter click.
Pro tip: Don’t try to capture everything. Start with 3-5 signals you know matter, and add more only if they actually help.
Step 2: Get Your Data Into Vitally
Automated health scores are only as good as the data behind them. If your CRM, product usage, or support data isn’t flowing into Vitally, your scores will be junk.
Common Data Sources
- CRM (Salesforce, HubSpot, etc.): Account owner, renewal dates, deal size
- Product Usage Analytics: Logins, feature usage, last activity date
- Support Platform: Ticket count, CSAT/NPS scores
- Billing System: Payment status, overdue invoices
How to Connect
- Vitally supports direct integrations with major tools (check their integration list).
- For stuff that isn’t covered, you can use their API or import CSVs.
- Make sure the data updates regularly—outdated info = misleading health scores.
Watch out: Integration setup can be tedious. Don’t be shocked if your data isn’t perfectly clean or mapped on day one. Fix the biggest gaps first, then iterate.
Step 3: Configure Custom Attributes in Vitally
You’ll need to turn raw data into usable fields (“attributes”) in Vitally. These are the building blocks for your health scores.
- Standard Attributes: Things Vitally pulls in automatically (e.g., Last Login Date).
- Custom Attributes: Calculated fields or unique data (e.g., “Has Used Feature X in Last 30 Days”).
Setting Up Custom Attributes
- Go to the “Attributes” section in Vitally.
- Create custom fields for each health signal you identified.
- E.g., “Product Usage in Last 7 Days” (True/False or a number)
- “Open Support Tickets” (number)
- “Payment Overdue” (True/False)
Tip: Boolean (yes/no) fields are easier to use in scoring logic than strings or free-text.
Step 4: Build Your Health Score Formula
This is where most people overcomplicate things. Resist the urge to make a complex, 10-factor algorithm—simple, transparent logic wins.
In Vitally: Health Score Models
Vitally lets you create multiple health score “models.” For most teams, start with one general customer health score.
Basic Example:
- Product Usage: 40% of score (active within last 7 days)
- Support Tickets: 30% (no more than 2 open tickets)
- Survey Feedback: 20% (NPS > 7)
- Billing Status: 10% (not overdue)
Set each attribute to contribute a percentage of the total score. Adjust weightings based on what actually matters for your business.
How to Set It Up
- Go to the “Health Scores” section in Vitally.
- Create a new score model.
- Add your attributes and assign weights.
- Define thresholds for “Healthy,” “At Risk,” and “Unhealthy.” For example:
- Healthy: Score > 80
- At Risk: 50–80
- Unhealthy: < 50
Don’t stress about getting it perfect. You’ll tweak these thresholds as you see real data.
Step 5: Automate Health Score Updates
The whole point is for scores to update automatically. Vitally recalculates health scores whenever the underlying data changes.
Make sure:
- Your integrations push fresh data at least daily.
- Custom attributes update automatically (not via manual entry).
- Everyone on your team can see the health scores on account records, dashboards, and reports.
Pro tip: Set up alerts or triggers in Vitally so the team gets notified when a customer drops into “At Risk” or “Unhealthy.” But don’t overdo alerts—otherwise, people will ignore them.
Step 6: Put Your Health Scores to Work
A health score is useless if it just sits there. Fold it into your team’s daily and weekly routines:
- Prioritize Outreach: Focus on accounts marked “At Risk.”
- Renewal Forecasting: Health scores should feed into renewal and upsell projections.
- Segment Customers: Build smart lists in Vitally for Healthy/At Risk/Unhealthy groups.
- Automate Tasks: Set up workflows (e.g., auto-create a task when a customer’s score drops).
Don’t: Use health scores as the only input for decisions. They’re a tool, not gospel.
What Usually Goes Wrong (And How to Avoid It)
Let’s be real—health scores often fail because teams set them and forget them. Watch out for these pitfalls:
- Overengineering: Complex formulas confuse everyone and get ignored.
- Bad Data: Garbage in, garbage out. Fix your sources before blaming the scoring logic.
- Ignoring Context: Not every “At Risk” flag means a customer needs a call. Use CS judgment.
- Stale Logic: As your product or customer base changes, revisit your score formula every few months.
Quick fix: Start simple, get feedback from your team, and adjust. Don’t be afraid to throw out signals that don’t help.
Bonus: Quick Wins and Things to Ignore
Do This: - Sync product usage and support data—those are usually the most telling signals. - Document your logic somewhere the whole team can see. - Review flagged “At Risk” accounts in weekly meetings.
Skip (for now): - Fancy machine learning or “AI-powered” scoring—most B2B teams don’t need it. - Tracking vanity metrics (e.g., how many webinars attended) unless you know they matter.
Keep It Simple, Iterate, and Trust Your Gut
Automated customer health scoring in Vitally isn’t magic—it’s just a tool to help you spot trouble (and opportunities) faster. Start simple. Use only the signals you know are meaningful. Review your health score formula every quarter or so, and don’t be afraid to simplify.
Most importantly: trust your team’s instincts alongside the data. No formula will ever beat a CSM who knows their customer.
Now go set up your health scores, get your team using them, and tweak as you learn. No one gets it perfect the first time—and that’s fine. Keep it useful, keep it honest, and you’ll actually get value out of Vitally.