UAE · ENBook a demo
IT CompanyCustomerInvestigator

Churn Prediction Agent

Flags accounts at risk of leaving.

Churn Prediction AgentCustomer Lifecycle
Industry
IT Company
Department
Customer
Team
Customer Lifecycle
Kind
Investigator
Approver
Account Manager
Code
CUS-14

It flags SaaS and managed-services clients likely to churn or not renew within the horizon, weighing usage decline, champion loss, repeated SLA misses, competitor RFPs, downgrade questions, a client's insourcing plans and late payments, and explains each flag with its evidence. It sets the churn risk on the account in the CRM and opens a save play for accounts worth it. **The problem it solves:** Why each account may leave or not renew: the signals present, how strong the evidence is, and whether our delivery or the client's situation drives it.

How it works

6 steps, and a person signs off before step 6.

  1. 01Looks it up

    Finds the record the request is about in the CRM: each account's ARR, renewal date, contacts and sponsor changes, open opportunities and activity notes..

  2. 02Reads the request

    Works out what is being asked and what a complete answer needs.

  3. 03Checks the facts

    Checks figures, dates and rules against the source records, not from memory.

  4. 04Works it through

    Weighs competing explanations against the evidence and shows which one holds up.

  5. 05Asks a person

    Hands the work to Account Manager with its reasoning before anything goes out.

  6. 06Acts

    Once approved, it acts in the CRM: it sets the churn risk on each account and opens the save play task. and records what was done.

What it works with

Your knowledge, your systems, your people.

Knows

The churn model: risk signals and their weights, past churn reasons by segment, save plays, and the ARR thresholds for each play.

Reads from

the CRM: each account's ARR, renewal date, contacts and sponsor changes, open opportunities and activity notes.

Acts in

the CRM: it sets the churn risk on each account and opens the save play task.

Hands back

A list of flagged accounts, each with account_id, probability_percent, arr_at_risk_usd, reasons (a ranked list with signal, evidence and weight), cause_side (ours, client or both) and save_play (or null).

Asks a person when

A flagged account is at or above 50000 USD ARR, or the cause is our own SLA or delivery performance.

ApproverAccount Manager

Settings

Your limits, set by your managers.

SettingDefaultWhat it controls
Flag at risk (%)60Accounts with a churn probability at or above this are flagged.
Horizon (days)120The window the risk is predicted for; enterprise renewals are decided months ahead.
Champion loss counts ashigh riskHow much losing the client's sponsor weighs.
Least ARR for a save play (USD)50000Accounts below this get a light-touch play instead.
Skip accounts younger than (days)90New accounts are left to onboarding.

Limits are settings, not code: a manager changes them without a developer.

Rules it always follows

  • Flag an account when its churn probability over 120 days is at or above 60 percent.
  • Treat loss of the client's sponsor as high risk; an issued RFP for the same scope is always high risk.
  • Explain each flag with the signals and their evidence, separating causes on our side (SLA misses, delivery slips) from the client's.
  • Open a save play for flagged accounts at or above 50000 USD ARR, owned by the account manager.
  • Skip accounts younger than 90 days.

See the Churn Prediction Agent agent on your own work.

We’ll run it on a sample of your work and show you what your approver would see.

Churn Prediction Agent · IT Company | Basinwright Industries