Churn risk investigated, not just scored
The Churn Risk Detector treats a falling health score as a question, not an answer. It correlates the signals, names the root cause, and raises only the alarms worth acting on.
Health score fell to 42 · Ridgeline
Churn Risk DetectorRidgeline’s health score dropped sharply. Correlating the signals before raising anything.
Correlating usage and login patterns4sLogins fell in the same week one user went silent, and that user was the champion.
Checking the champion’s current role3sShe left the company last month. Two frustrated tickets from her old team followed. This is real risk with a nameable cause.
Alerting the CSM with a save plan1s
How it works
Health score fell to 42Logins down 40%Ticket escalatedSponsor changed · Meridian
Save plan to CSMRoot cause documentedRisk record updatedExec check-in proposedTriggered by moving signals
A health score drop, a login decline or an escalated ticket in your CRM or support desk opens an investigation the same day, on any account.
Correlates signals into a verdict
Usage, logins, ticket sentiment and sponsor activity get weighed together. The agent names the root cause, or says honestly that the signals do not add up to one.
Delivers the plan to your team
Real risks reach the CSM in Slack with a save plan matched to the cause, and the CRM record carries the diagnosis and the evidence.
Without Relevance
With Relevance
A health score turns red and nobody knows what it means beyond "look into this".
Every flag arrives with its root cause named: the champion who left, the integration that broke, the team that went quiet.
Logins are down, so the account is "at risk", even if the product is being used more than ever through the API.
Contradictory signals get investigated together, and the conclusion follows the whole account, not the single scariest chart.
One angry ticket triggers an escalation, and after the third false alarm the team stops reading them.
A single bad signal on an otherwise healthy account becomes a watch item, not a fire drill. When the agent escalates, it means it.
The playbook is the same for every red account: schedule a call, offer a discount.
The plan matches the diagnosis: rebuild sponsorship when the champion left, fix and apologise when the product failed, re-onboard when adoption stalled.
Risk gets a human read on the accounts large enough to justify one.
Every account gets the same investigation the day its signals move, from the largest logo to the smallest.
The model outputs 42 and nobody can say why, so nobody quite trusts it.
Every conclusion shows its evidence, and when the signals do not add up to a verdict, the agent says that instead of forcing one.
Held to a quality bar, on every run
Every risk call is checked against eval test cases written for this exact job before your team acts on it. If a run fails the bar, it never ships.
Churn Risk DetectorPass rate · last 14 days
Sampling 2% of live runs · 97% passing this week
Train your agent like an employee
Onboard it with your playbooks and correct it in plain English. It learns the lesson for good, and nothing changes until you approve it.
Built for enterprise teams
Run agents on your real data with the access controls, audit trails, and residency guarantees enterprises require all built in.
Monitoring
Real-time visibility into every agent’s activity, performance, and cost.
RBAC
Control who can use, build, edit, and deploy agents.
Data residency
Multi-region deployment keeps your data within your required geography.
Version control
Full version history on every agent. Roll back to any previous state.
Audit logs
Every action, every decision, every tool call logged and exportable.
Human-in-the-loop
Set approval gates on any action or allow agents to ask questions.
SSO / SAML
Enterprise single sign-on with SAML 2.0. Centralize identity & access.
PII masking
Detect and redact personally identifiable information.
"The ability to be vendor agnostic and the ability to scale across a breadth of functions is a really key feature."
Levi Watters
Partner, KPMG Australia
"The key for us was how we can modularize industry knowledge and the best playbooks, and apply it."
Allen Roh
Senior Marketing Manager, Autodesk
"We're looking for every place where AI can allow sellers and customer success reps to be more engaged with customers."
Rob Giglio
Chief Customer Officer, Canva
"The ability to be vendor agnostic and the ability to scale across a breadth of functions is a really key feature."
Levi Watters
Partner, KPMG Australia
Team it up with more customer success agents
Deal won- 1Handoff brief sent
- 2Intro email sent
- Kickoff booked
Customer Transitioner
Hands new customers off from sales to success.
Onboarding Guide
Drives new customers through every setup step.
UpsellUpsell & Cross-Sell
Spots upsell and cross-sell potential per account.
Every agent you add shares one stack
These agents all run on one platform: one gateway, one router, one eval suite, one audit trail. Every agent after the first ships faster.
- Triggered by key events or signalslike Zapier
- Given access to contextlike Zep
- Connected to your appslike Composio
- Access to all LLMslike OpenRouter
- Performance evaluatedlike Braintrust
- No-code agent builderlike Dust.tt
- Coordinated into teamslike CrewAI
- Chained into multi-step processeslike n8n
- Kept alive through long-running worklike Temporal
- Traced at every step of every runlike Langfuse
Put the Churn Risk Detector to work this quarter
We partner with enterprise teams to take this work off their plate. An agent built with you, proven against your quality bar, and live in weeks.