As per me, Analytics Projects get into these four broad phases-
- Business Problem Phase– What needs to be done?
- Increase Revenues
- Cut Costs
- Investigate Unusual Events
- Project Timelines
- Technical Problem Phase – Technical Problems in Project Execution
- Data Availability /Data Quality/Data Augmentation Costs
- Statistical -(Technique based approach) , Hypothesis Formulation,Sampling, Iterations
- Programming-(Tool based approach) Analytics Platform Coding (Input, Formats,Processing)
- Technical Solution Phase – Problem Solving using the Tools and Skills Available
- Data Cleaning /Outlier Treatment/Missing Value Imputation
- Statistical -(Technique based approach) Error Minimization, Model Validation, Confidence Levels
- Programming-(Tool based approach) Analytics Platform Coding (Output, Display,Graphs)
- Business Solution Phase– Put it all together in a word document, presentation and/or spreadsheet
- Finalized- Forecasts , Models and Data Strategies
- Improvements in existing processes
- Control and Monitoring of Analytical Results post Implementation
- Legal and Compliance guidelines to execution
- (Internal or External) Client Satisfaction and Expectation Management
- Audience Feedback based on presenting final deliverable to broader audience
Looks like CRISP-DM phases that can be extended with your vision of activities to be taken . Thanks …
In the second phrase, what is the general rule for data quality evaluation?