Establishing Enterprise Data Standards - High-Performance Web Databases: Design, Development, and Deployment - page 19

deliverables (e.g., screen images) can also be included in the repositories
to provide project teams with the ability to get started on their projects
quickly. Paper-based, or CD/ROM-based, delivery of the standards and tem-
plates should accompany the electronic delivery modes.
MEASURING COMPLIANCE AND ENFORCING STANDARDS
During the pressure of development projects or unexpected operational
emergencies, it is not uncommon for all standards, including data stan-
dards, to be suspended until the immediate problem is resolved. In some
instances, this is acceptable in the short term. However, compliance is
important in the long term, and several mechanisms should be used to
measure and enforce this, as follows:
• Deliverable walk-throughs. Regular sessions should be scheduled to
walkthrough and sign off on project deliverables to ensure that they
comply with the published enterprise data standards. DBAs should be
involved in this process.
• Audit. Infrequent audits of projects across the enterprise should be
used to ensure ongoing compliance with the published standards. Re-
sources on the audit teams can vary over time. A good size for the au-
dit team is about three resources. It is a good idea to rotate resources
on and off the teams over a period of time.
• Enforcement. Examples of noncompliance should be documented and
the appropriate project manager(s) should be mandated to ensure
that the standards are adhered to within a specific timeframe (e.g., one
month).
• Quality plan. Every project plan should be accompanied by a quality
plan that includes the activities that will be followed by the project
teams to ensure compliance with the standards.
TEN TIPS FOR GETTING STARTED
Many projects have shown that the following tips greatly simplify the
process of establishing enterprise-wide data standards:
1. Keep them simple.
2. Borrow, if possible, avoid creating.
3. Use samples from actual projects.
4. Make the standards readily accessible electronically.
5. Do not hesitate to modify enterprise data standards if a good argu-
ment is presented.
6. Use standards as reasonable guidelines.
7. Build early successes as showpieces.
8. Build data standard compliance right into the project plan or quality
plan.
9. Never create standards in a vacuum.
10. Enforce the use of the data standards.