Play 5 – Get Data In Shape
- Difficulty: Moderate to Hard
- Target time: 12-24 weeks
- Maturity: 2 to 3 3 to 4/4.5
- Inconsistent data in the organisation Management Information is often wrong
- Teams have no idea where to source data
- Databases known
- Applications known
- Areas of biggest concern known
- Processes using & creating data broadly known
- No overall data model
- Data mastered in multiple places
- Lots of differing opinions on data and definitions (which makes data hard to manage)
- Communication to relevant IT teams holding data from a senior leader
- Business support agreed
- An agreed overall nominee to make the final decision where disagreement
- An initial area of focus is agreed
- Create an data catalogue for key data items in the organisation. Agree the definitions with the business
- Communicate the data dictionary for projects to utilise in design
- In parallel, gather the data stores that hold the data and the applications that use those data stores (just worry about the key data, not everything)
- Identify the applications mastering data and show where multiple masters exists and outputs that use that data
- You need to identify the data masters, but you will need to understand your processes to do that effectively (not in this play). Once data masters are understood then projects can source their data from the correct applications
- Capture data flows into and out of applications
- Consistency of data definitions helps projects with their data modelling
- The mapping to applications helps speed project delivery with information flow understandin
- Understanding sources of data ensures that MI can be delivered with knowledge as to the source of data (system or manual). If you overlay data quality scores then you can also measure the quality of reports. This often leads to initiatives to remedy data quality issues
- CIO – key areas of concern
- Business Owners – data definitions
- Application Owners – application information and dependencies
- DBAs – data stores supporting applications and data stored
- Engage and hold regular updates with the CIO
- Share data catalogue and dependencies with project teams
- Once applications that master data are identified and the reports/business outputs are understood, communicate areas of business risk, e.g. “sales report is missing key data for regions X, Y & Z?” (this is a real example from a CEO report)
- Engagement across the business, and reduction of data errors, brings EA credibility
- IT projects engage EA leveraging the enterprise architecture data artefacts to speed project delivery
- Exposure to senior management using MI means senior awareness of capabilities
Steps | |||
|---|---|---|---|
What | Capture the Data Subjects and associated Data Objects with descriptions | Identify the locations of the databases/data stores that contain the Data Object | Capture the applications with basic details and services |
Usage | Capture the Data Subjects and associated Data Objects with descriptions | Understanding of where data stores are. This can later be an anchor for identifying where sensitive data is stored and what it is being used for (when we tie it to Applications and processes, e.g. for data privacy analysis) | Have a list of applications used that can be the anchor for the rest of the work. Define strategic applications in line with your target state |
Interested Parties and expected response | Project Teams | CIO | Business & IT |
Approach | Overall Enterprise | By Critical Application or By Business Area | By Critical Application or By Business Area |