Predict business impact monitoring content

If you use Guidewire Predict, Explore includes Liveboards with metrics that help you monitor how your predictive models impact business key performance indicators (KPIs). These business impact monitoring (BIM) Liveboards can help you prove the value of the models or discover where they need improvement. For example, to quantify the value of a claims predictive model, you could monitor the average amount paid on claims or the average days to close, before and after the model was implemented.

To learn how to create your own BIM Liveboards, see Monitor the impact of Predict models in Explore Liveboards.

Required applications and services

To use BIM Liveboards, your organization must already have the following Guidewire applications and services in Guidewire Cloud:
  • Predict, with at least one model deployed and sending scores to Analytics Manager.
  • Analytics Manager, with a solution enabled to mediate between Predict and an InsuranceSuite core application.
  • An InsuranceSuite core application, such as ClaimCenter, with feature flags enabled to interact with Analytics Manager. (See the ClaimCenter Configuration Guide.)
  • Data Studio
  • Explore

Lineage of Predict BIM content

The following diagram shows the lineage. Click to enlarge the image.
Solution datasets in Data Studio are published to Explore and become tables, then models and liveboards.
  1. Guidewire created Predict BIM solution datasets that live in Data Studio. Each solution dataset joins at least two datasets:
    • A dataset that contains InsuranceSuite KPIs
    • The assessment summary dataset that contains Predict scores and assessments
    You can find details about the solution datasets in the Data Studio Data Dictionary. (See the Data Studio guide.)
  2. The solution datasets are then published to Explore and used to create models. There's one Explore model for each dataset.
  3. The models are the data source for Liveboards.

Using BIM Liveboards

At the top some BIM Liveboards, there's a filter that allows you to choose a specific predictive model, such as Segmentation. You must use the filter. Without filtering, you’ll see duplicated data because there might be multiple predictive models in the data source. For example, a single claim might have two rows in the data source: One row with a score from a segmentation model, and another with a score from a subrogation model.

BIM Liveboards

Liveboard name Key contents Data source (Explore model)
Business Impact Monitoring - Claims
Note: You must have a claim-level model in Predict.

Includes analysis of how claim-level Predict models impact claims KPIs. Compares claims incurred before and after the deployment of a Predict model. Helps claims executives, managers, and actuaries prove the value of claim-level Predict models or discover where they need improvement.

 BIM Claim Model Results
Submission Business Impact Monitoring
Note: You must have a submission prioritization predictive model in Predict.

Includes analysis of how a submission prioritization predictive model impacts KPIs for bound policies, such as bound policy count, quote to bind ratio, and quote to bind days. For example, the Liveboard counts the number of bound policies that are predicted to be profitable or have a high loss ratio. Then, it compares the counts before and after the deployment of the Predict model. This helps underwriting executives, managers, and actuaries prove that a submission prioritization predictive model leads to more efficient and profitable underwriting, or discover where the model needs improvement.

 BIM Submission Model Results

BIM Models (data sources)

Note: Here, an Explore model refers to the data source in Explore that's used to build the final Liveboard.
For more details, see Data dictionaries for Explore models.
Explore model name Key contents Source solution dataset from Data Studio
BIM Claim Model Results

Includes all claims and their scores and assessments from claim-level Predict models.

Helps Predict users prove the value of claim-level predictive models or discover where they need improvement.

 bim_claims_model_results
BIM Submission Model Results

Includes details about submissions, the submission process, and scores and assessments from a submission prioritization Predict model.

Helps underwriting executives, managers, and actuaries prove that a submission prioritization predictive model leads to more efficient and profitable underwriting, or discover where the model needs improvement.

 bim_submission_prioritization