Skip to main content
Signal AI reports an accuracy range, not a single score — the range’s width reflects how precise future predictions are expected to be.
Your Signal AI insights improve by learning from the data in your own calls. When you provide training data, the predictive power of that Signal is reflected in your Signal AI accuracy score.

​
Viewing your Signal AI accuracy score

  1. Log in to your Invoca account. From your network-level dashboard, select Signal in the gray menu ribbon.
  2. In the list, select a Signal with the type “AI (Custom).”
In the Live Predictive Model tile, your Signal AI is measured with an accuracy range instead of a single score. This range shows how well your Signal AI is categorizing calls within the training data you’ve provided, and how well it’s expected to perform on incoming, unclassified calls once deployed.
  • The percentages at either end of your accuracy range show how accurately your Signal AI model predicted different outcomes in your training data.
  • The size of that range correlates with how confident we are in the precision of those predictions going forward — a narrower range means better precision.
Showing both measurements helps represent how your Signal AI predictions improve as you provide more training data. For example, imagine you create a first version of your Signal AI using 500 calls for training data. It learns from 400 of those calls and sets aside the other 100 to test the pattern it finds. Of those 100 calls, it accurately predicts all 100 outcomes — a 100% accuracy score. You then add another 750 calls to update your Signal AI to a new version, but in that next sample, it only predicts 700 out of 750 correctly. Your Signal AI didn’t get less accurate from version to version — it used the new data to recognize stronger patterns. Your first sample may have been more uniform or easier to predict than your second, resulting in a lower accuracy score but more precise predictions overall.

​
Where to go next

Last modified on September 23, 2026