Defining the Problem
What problem are you trying to solve?
What are the main impacts of the problem?
What predictions or recommendations are being made?
Identify prediction variables (X) and objectives (Y)
What data sources are available?
Is there enough data? Can it be used?
What are the appropriate models according to the predictions/recommendations described in item 2?
List the most relevant ones.
How will model performance be evaluated?
Define the success criteria.
What transformations will be required to tailor the available data to run the chosen models and achieve the expected outcome?