Skip to main content

Predictions

A Prediction is an account-level score that estimates the probability of churn or renewal — calculated continuously from your actual customer data, not a manually configured formula. Where health scores ask you to decide in advance what matters, FunnelStory's prediction models learn the patterns from your own historical outcomes: which accounts renewed, which churned, and what their data looked like in the months before.

The result is a score grounded in the specific reality of your customer base, not a generic industry template.

The Health Score​

Every account receives a health score from 0 to 100: 50 is neutral, below 50 is increasing churn risk, above 50 is a healthy trajectory toward renewal. The score is a net result of two competing signals — the probability the account will stay, weighed against the probability it will churn — so an account with high product usage but also high support escalations will land near the middle until the pattern resolves.

Alongside the score, every prediction carries a predicted outcome (Churn / Retention / Neutral) and a confidence level (High / Medium / Low / Neutral). See Predictions overview and Confidence Ratings for the full breakdown of both.

Driving Factors and What-If Analysis​

Each prediction surfaces the specific factors moving the score, split into what's currently supporting retention and what's pushing toward churn — each shown relative to your broader account population so it's clear whether an account is above or below average on any given signal. What-If Analysis lets you simulate how changing one of those factors would move the score, useful for prioritizing what to address before a renewal conversation.

See Driving Factors and What-If Analysis for the full detail, including how factors pull from both structured and unstructured data.

How Predictions Learn Your Business​

FunnelStory's models train on your own historical outcomes — which accounts renewed, which churned, and what preceded each — rather than a generic baseline. You define those outcomes through Revenue Tags, and each Needle Mover type carries a configurable impact weight controlling its influence on the score. The model retrains as new outcomes are recorded, so it keeps adapting to your business over time.

See How Predictions Learn Your Business for the full detail on Revenue Tags, Needle Mover weights, and the retraining loop.

Per-Product Predictions​

Accounts with multiple products get a separate health score and driving-factors breakdown per product line — useful when products have different renewal timelines, separate CSM ownership, or you need to isolate which product relationship is at risk. See Product Predictions.

Acting on Predictions​

Predictions are designed to trigger action, not just inform awareness. See Acting on Predictions for the full list of what you can do from any account's prediction view — reviewing driving factors, running a What-If analysis, jumping to Needle Movers, pushing a CRM task, or asking Renari.

Relationship to Needle Movers​

Predictions and Needle Movers are complementary, not redundant.

A Prediction gives you the score — the probability that an account will churn or expand. A Needle Mover gives you the reason — the specific, sourced signal (a competitor mentioned in a QBR, a champion who has gone quiet, an unresolved pricing concern) that is moving that probability.

Together they provide both the "what" and the "why" needed to take confident action.

  • Predictions overview — full operational detail: health score, confidence, driving factors, revenue tags, and acting on predictions
  • Needle Movers — the specific signals driving prediction scores
  • Customer Intelligence Graph — how prediction scores are computed and stored as derived intelligence
  • How FunnelStory Works — where predictions fit in the pre-computed intelligence layer
  • AI Agents — automating responses when predictions cross risk thresholds