Database Reference
In-Depth Information
Figure 12.4
Analytics plan for YoyoDyne Bank case study
In addition to guiding the model planning and methodology, the analytic plan
contains components that can be used as inputs for writing about the scope,
underlying assumptions, modeling techniques, initial hypotheses, and key findings
in the final presentations. After spending substantial amounts of time in the
modeling and performing in-depth data analysis, it is critical to reflect on the
project work and consider the context of the problems the team set out to solve.
Review the work that was completed during the project, and identify observations
about the model outputs, scoring, and results. Based on these observations, begin
to identify the key messages and any unexpected insights.
In addition, it is important to tailor the project outputs to the audience. For a
project sponsor, show that the team met the project goals. Focus on what was done,
what the team accomplished, what ROI can be anticipated, and what business
value can be realized. Give the project sponsor talking points to evangelize the
work. Remember that the sponsor needs to relay the story to others, so make
this person's job easy, and help ensure the message is accurate by providing a
few talking points. Find ways to emphasize ROI and business value, and mention
whether the models can be deployed within performance constraints of the
sponsor's production environment.
In some organizations, the data science team may not be expected to make a full
business case for future projects and implementation of the models. Instead, it
needs to be able to provide guidance about the impact of the models to enable the
