Learn how to activate your scoring model, A/B test multiple scoring models, and compare results.
- ACTIVE: activate the scoring model immediately and route the information back to your primary source.
- TESTING: activate your scoring model in testing mode. This will not sync your lead score to your primary source.
- DRAFT: save your scoring model as a draft and activate it later.
Note: you can edit your scoring model at any time. To edit or change the scoring model type, navigate to the Scoring Models page from the top navigation. From here, click on the name of the model you wish to edit.
How to save your scoring model as a draft
If you're not ready to activate your scoring model immediately, you can save it as a draft and come back later to edit and/or activate it.
How to activate your scoring model
You can immediately activate your scoring model and route the information back to your primary source.
In this case, you'd also need to choose which fields to fill for the scores.
In this example with HubSpot, you can choose to use the defaults that are listed here or you can select existing fields in HubSpot by clicking "Use existing fields" in the top right-hand corner of the field mapping section.
Note: Depending on the primary source you are using, it may not be possible to use custom fields. Please visit our Integrations page for more info on what each supported primary source can do in Breadcrumbs.
How to A/B test your scoring model and compare results
If you are not ready to write the scoring information back to your primary source but are still interested in learning how your model is performing or if you want to test different scoring models, you can use Breadcrumbs' testing mode.
When doing so, the model still retrieves data to analyze it but it will not add the lead scores back to your primary source.
This way you'll be able to analyze how different scoring models perform and optimize your scoring model.
How to compare and analyze the results of your A/B tests
In order to understand the results of your A/B test, all you have to do is navigate to Tools > Compare.
Here, select up to 4 different scoring models either in testing or active mode, and see which one is performing better in terms of co-dynamic distributions, fit categories, and activity categories.