2. Information
Value
$0
Data Quality
Saturday, May 16, 2009
3. Poor quality customer data costs U.S.
companies $611 billion annually in postage,
printing and staff expense.
- The Data Warehouse Institute
The cost of poor data quality can reach as
high as 15% to 25% of operating profit.
- The Data Warehouse Institute
At least 25% of critical data within Fortune
1000 companies will be inaccurate.
- Gartner
Saturday, May 16, 2009
4. Billing and payment errors cause negative
customer perceptions and affect a
company’s ability to accurately state their
financials
Operating expenses are inflated due to
returned mail, and the manual rework to get
it sent correctly
Regulatory fines are levied due to inaccurate
reporting of data to government entities
Saturday, May 16, 2009
5. Customers (and revenue) are lost due to an
inability to track customer interactions or to
recognize high-value customers
Negative publicity is generated when a
company fails to meet customer obligations
on a large scale, like a disruption of service
Flawed analytics lead to poor tactical and
strategic decisions
Saturday, May 16, 2009
6. Extra time is required on IT projects to
reconcile data
Delays are incurred in deploying new
systems
Credibility in a system or application is lost
when it doesn’t perform as advertised
Saturday, May 16, 2009
7. Completeness – is all relevant data entered?
Consistency – is the data entered in the
same format?
Accuracy – is the entered data correct?
Relevance – is the data collected useful?
Timeliness – is the data available when
needed?
Integrity – is the data consistent when
duplicated?
Saturday, May 16, 2009
9. Assess
Improve
Data Cleansing
Prevent Data Quality Deterioration
Recognize Data Imperfections
Monitor
Saturday, May 16, 2009
10. Claims Data Underwriting Data
Many data quality edits Several underwriting
and rules built into and policy admin
CWS systems with various
Data Quality levels of DQ edits
Scorecard in place; No data quality
most quality is good scorecard in place
Targeted projects as Some data clean-up
needed to cleanse associated with other
data projects
No metadata for data No metadata for data
exceptions exceptions
Saturday, May 16, 2009
11. Tasks
Begin assessment of underwriting data
Document data quality rules
Begin data cleansing as possible
Challenges
No dedicated business resources
Limited IT resources
No automated cleansing tools
Saturday, May 16, 2009
12. Tasks
Review and update the Corporate Data Strategy
Analyze/evaluate resources needed to support a
more comprehensive enterprise data quality
program
Secure support for enterprise data quality
program
Challenges
Competing business priorities
Limited corporate resources
Saturday, May 16, 2009