Database Reference
In-Depth Information
as the effects of key variables on predicting churn, and the relationship of key
variables to other variables. The team may even want to make suggestions to
improve the model, highlight any risks to introducing bias into the modeling
technique, or describe certain segments of the data that may skew the overall
predictive power of the methodology.
12.2.8 Recommendations
The final main component of the presentation involves creating a set of
recommendations that include how to deploy the model from a business
perspective within the organization and any other suggestions on the rollout of
the model's logic. For the Yoyodyne Bank example,
Figure 12.15
provides possible
recommendations from the project. In this section of the presentation, measuring
the impact of the improvements and stating how to leverage that impact within
the recommendations are key. For instance, the presentation might mention that
every customer retained represents a time savings of six hours for one of the
bank's account managers or $50,000 in savings of new account acquisitions, due
to marketing costs, sales, and system-related costs.
Figure 12.15
Sample recommendations for a data science project
For a presentation to a project sponsor audience, focus on the business impact
of the project, including risks and ROI. Because project sponsors will be most
interested in the business impact of the project, the presentation should also
provide the sponsor with salient points to help evangelize the work within the
organization. When preparing a presentation for analysts, supplement the main set
of recommendations with any implications for the modeling or for deployment in a
