Swallow
  • YOUR FIRST MODEL
    • Setting up your first project
    • Manual setup of a new product
    • Excel Model Upload
    • Create your team
    • Essential concepts
      • Syntax
      • Steps
      • Collections
      • Try Model
      • Spotlight
  • PROCESS
    • Team Management
    • Governance & Collaboration
  • BUILD
    • Building with AI
    • Step builder
      • Understanding step types
      • Ordering step blocks
      • Placeholder mode
      • Up/Downloading models
      • Sharing your model
      • Running tests
    • Inputs
    • Enriching Raw Data
      • Transform Data
      • Enrich via API
      • Data Set
      • Collection
      • Custom Code
    • Setting Up Decisions
      • Exclusions
      • Excesses
      • Endorsements
      • Refer
    • Making Calculations
      • Factors
      • Calculation
    • Outputs
  • TEST
    • Creating New Tests
    • Running Tests
    • Editing Tests
    • Reading & Interpreting Tests
    • Impact Testing
  • DATA
    • Working with real-time data
    • Data Views
    • Data Quotas
    • Troubleshooting
    • API Webhooks
    • Data Downloads
  • ADMIN & SETTINGS
    • Project Management
    • Account Management
  • INTEGRATIONS
    • Introduction
    • Payments
      • Stripe
      • GoCardless
      • Adyen
    • Alerting
      • Slack
      • Zapier
    • Data
      • AWS S3
      • Google Sheets
    • Admin
      • Socotra
      • Guidewire
      • Hubspot
      • Salesforce
  • Developer API
    • Introduction
    • Authentication
    • Errors codes
    • Endpoints
      • /quotes
      • /projects
      • /tests
      • /process
Powered by GitBook
On this page
  1. TEST

Reading & Interpreting Tests

Using testing reports to analyse your model's performance

PreviousEditing TestsNextImpact Testing

Last updated 1 year ago

Once you've built out your test cases, you can run all of your tests to see the impact your current model is making.

You'll be able to compare the difference between your actual results and the outputs you'd projected, as well as see the spread of prices so you can refine your pricing strategy.

The avergage result value will indicate what the average quote that is served to customers will be. The validity shows what the likelihood of a valid (non-excluded) quote being shown.

Movement shows the change between what the expected result (which could be the previous versions result) against the new pricing changes, which will give you an indication of impact.

Reporting in the test tab