Framework for executing data projects

Following a framework when executing a data project could help reduce downtime, plan ahead, and have an established process in the business others could follow.

Framework

  • Investigations
    • Catch up with the business requirement (BR) solicitor 
    • BR documentation by the solicitor
      • Background
      • Previous process 
      • Key links to reference documents 
      • Acceptance criteria 
    • Understand previous process and value to be created
    • Check data sources and documents for previous or similar processes 
    • Data quality observations
    • Data security
  • Solutions Implementation
    • Create data analyst project documentation 
    • Identify sources for new process
    • Catch up with BR sponsor/solicitor to manage expectations
    • Create new process
    • Reconciliation/Change management
      • Hypothesis around issues or success derived from new process
      • Extra insights generated from new process
      • Confirming or discarding hypothesis with solicitor or stakeholders
    • Document new process
    • Data governance

Catch up with the BR sponsor/solicitor 

  • Get an overview of the value to be created
  • Get overview of areas of the business that will be impacted

BR documentation by solicitor

  • Having a worded request written down could immensely enhance understanding and remove possible grey areas. The documentation should contain:
    • The background of the request
    • How the previous process was carried out? Who does what?
    • Link to the reference documents on the previous process
    • A clearly detailed acceptance criteria for the solution to be created 

Understand previous process and value to be created

  • Aggregating knowledge gathered from the catch-up and BR documentation by the solicitor to get an overview of the value created by the previous process and the potential value the new process could create 
  • Meeting with stakeholders around the BR to get a more detailed understanding of the process

Check data sources and documents for previous or similar processes 

  • What system or systems was used in capturing the data used in the previous process
  • Where is the single or multiple source of truth data for the previous process

Data quality observations

  • Is the data entered accurately, are there missing values, how frequently is the data updated?
  • Does the business logic correlate with the logic used for the features
  • Catch up with stakeholders to discuss observations

Data security

  • How was the data secured? Who had access to the data? Are the people having access to the data meant to have access to the data?
  • Can a similar or better security be replicated using a new process

Create data analyst project documentation

  • Create a new documentation that will be more detailed to the execution of the project. Use the documentation from the BR solicitor as a starting point.
    The documentation could contain:
    • Background 
    • Previous process
    • Key links: Project tracking tool (e.g. Jira or Asana) | Previous process related documents | New process related documents | Data governance | Background reference
    • Data requirements containing the features and business logic required to achieve the BR
    • Key observations
    • Reconciliation/Change management
    • Acceptance criteria
    • Activity notes (if required)
  • Create a ticket on a project tracking tool (e.g. Jira or Asana) 

Identify sources for new process

  • Are the sources of data only from systems, or just manual, or system + manual
  • Can a direct connection be made from the system to the BI Tool so the data can be ingested directly from the system
  • Can a power app be made for manually entered data so the data can be ingested from the power app

Catch up with BR solicitor to manage expectations

  • Meet with the BR solicitor to discuss based on your investigations, what can be achieved and what cannot at the point in time. This could be a continuous conversation.

Create new process

  • Start building the new process using the available data analytics tool that can be better used for the process

Reconciliation/Change Management

  • Is the values from the newly built report equal to the previously built report
  • Is the values from the newly built report different from the previously built report due to data quality issues
    • Are the definitions of features from the newly built report what was intended by the business
  • Catch-up with stakeholders to discuss usability and impact of new process for the BI report

Document new process

  • Document the new process for generating the report using the BI tool

Acceptance criteria 

  • Newly built report on BI tool generated same or better report than spreadsheet 

Data governace

  • Define similar or new data governance measures that have been put in place

Codes

  • Document codes similar to that used on the spreadsheet report that were use on the BI tool
  • Document new codes for new insights where applicable

Framework for executing data projects Read More »