Operationalizing Data Quality through Data Governance - Data Governance: Creating Value from Information Assets - page 68

across-dataset relationships as well as within-dataset relationships. There
are two levels of traceability:
1. Metadata traceability enables the user to understand the relationship
of the analysis variable to its source dataset(s) and variable(s) and is
required for derived data compliance. This traceability is established
by describing (via metadata) the algorithm used or steps taken to
derive or populate an analysis value from its immediate predeces-
sor. Metadata traceability also is used to establish the relationship
between an analysis result and analysis dataset(s).
2. Data point traceability enables the user to go directly to the specific
predecessor record(s) and should be implemented if practically fea-
sible. This level of traceability can be very helpful when a reviewer is
trying to trace a complex data manipulation path. This traceability
is established by providing clear links in the data to the specific data
values used as input for an analysis value.
None of the data quality dimensions is complete by itself, and, frequently,
dimensions are overlapping. Data quality is not linear; having data quality
on one dimension is as good as “no quality.”
IMPORTANCE OF DATA QUALITY
Information/data penetrates into corporations' networks, systems, and
storages, thus delivering volumes of data about customers, products,
partners, and new opportunities. After analysis and review of that infor-
mation is complete, data then flows back out in the form of new prod-
ucts and services, documents, and applications. Organizations invest
heavily in mission-critical business software initiatives trying to cap-
ture each drop of information as it cascades into data centers around the
clock. However, data typically resides in multiple sources with various
data structures. There are inconsistent rules for changing, improving, or
accessing it. No one can agree on which data quality measures should
be applied; business users and IT personnel view information differently
and do not speak a common language. To make matters worse, many
companies use disparate, poorly integrated, data quality management
tools and applications.