EIM as disparate as it may sound has many definitions; Here I plan to just publish just the main causes of data being considered as not useful!
To provide jazzy reports and ultimate analytics in a complex landscape always ends as useless unless it has a solid foundation, unless the information contained in it can be trusted ;
• Integrate information from various different sources
• Provide consistent business definitions, so that everybody understands and agrees on the meaning of the KPIs used
• Provide information in a timely manner, so users can be alerted to problems in time to be able to do something about them
• And, perhaps above all, you need to provide accurate, quality information
Brand new survey I read recently on the web said that up to 75% of information workers have made business decisions that later turned out to be wrong due to flawed data, that only 10% of them always have all the information they need to make business decisions, and that they spend up to 30% of their time verifying the quality of the data they use.
One will always have multiple systems. Of course a client can—and should—combine some of the information, to make data marts and data warehouses, but some important business data will always be missing from those systems. The only hope is to use a flexible combination of ETL and BI techniques to provide the appearance of a coherent, single version of the truth. The key here is to use multiple techniques, choosing the right approach for each source and type of data, and to keep track of all the metadata so that you have a coherent, consistent approach.
So what types of data quality problems can companies expect to find? For example,
• Are the address fields 50% empty?
• Have the customers moved recently at all?
• Do you have a large percentage of default address customers?
Many of these are simply data entry errors.
Some Clients may also have errors that fall under the category of data inconsistency. These errors are typically violations of defined data standards.
The ETL process can introduce data quality errors as well. For example, it is very typical for data conversion to require normalization, or standardization, of data values or data formats. ETL process doesn’t account for the fact that one can likely have information stored for the same customer in various formats in multiple source systems, you could end up with a single customer showing up 3, 4 or even more times in the SAP data warehouse. Example : John Miller / Miller John / Miller A John can be the same customer maintained in three different systems.
So what does Data Quality Offer a Client ?
- Data Quality improves decision making,
- It provides a single, accurate view of information and complete view of data across your enterprise
- It ensures that compliance reporting is accurate (which allows the executives to rest a little easier)
- It enhances targeted marketing efforts and ensures that you avoid customer service faux pas.
- It provides new insight into business opportunities to reduce costs in areas such as supply chain optimization.
Thus this is a summary of the problems one would see in providing a solid foundation for trusted BI. Data quality, disparate data, inconsistent terms, and traceability back to the source.
EIM will improve service delivery in –>
• OPS-wide modernization
• Faster to find information
• Fact-based decision-making
• Less chance of duplication
• Collaborate across organizations
• Search & share information
• Workflow processes
EIM will reduce risk by ;
• Know who owns each information resource
• Apply access permissions to protect privacy & sensitivity
• Find information
• Requested under Freedom of Information
• For legal e-discovery
• Keep versions and audit trail
• Transfer or destroy records when no longer needed
• Protect organizational memory