Home » Business

Category Archives: Business

(TechNeed) Master Data Solutions for Business Intelligence

Within the Business Intelligence (BI) space we are trying to make better decisions more quickly by integrating data from different feeding systems into one BI model. There are a number of steps involved that are depicted here in a simplified way:
1) Data feed from feeding Systems
2) Map & Match to master date
3) Storing in Warehouse
4) Extracting & integrating
5) BI Analytical systems
This Need focuses on Step 2.
Data is fed from internal and external systems, and is typically not easy to align to one set of master data. Common issues we face:
- Data Standards differences
– Dimensions, attributes and hierarchies have different definitions both internally and externally
- Granularity and timing differences
– some data is by product code, other data is at brand level
– some data is by month, other data is by day
– much data is pre-aggregated, limiting flexibility
– snapshots of data exist at different points-in-time across data types
- Dimensional differences
– Some feeding systems contain up to 8 dimensions, others have 3 dimensional models
- Scope differences
– Internal data only contains Company business, external data can contain competitor data
- Format differences
– Some data comes from mainframe systems, some is received as flat files or even Excel Files, all with different file format (simple headers, complex headers, Unix file types and even EDI formats)
- Language differences
– Description in products, customers, time etc. can appear in multiple languages.

In order to align data to one set of master data across all sources, we are looking for solutions that allow us to map and align data easily across multiple sources on multiple different platforms. Once mapped to a single BI standard, data is easier to integrate or store in our BI data warehouses.

Date published: 18 October 2012

Follow

Get every new post delivered to your Inbox.