Karen Hall, the district's data steward had high expectations for the new metadata system. But the astute data expert noticed something strange happening during installation, and she began to worry.
As Karen helped the vendor map the district's data elements and data sets to the new system, she identified numerous mistakes in her data's format, structure, and logic. Invariably, the vendor had the same response whenever a problem presented itself: "You don't want us to deal with that... It would take way too much time to straighten it out and would delay the implementation of the new system."
The first couple of times this happened, Karen kept a mental note of what needed to be corrected in the system, assuming she would fix the problems at a later date. But by the time the list had grown too long to trust her memory, she decided to raise the issue more formally with the vendor.
She was very direct: "So how will this metadata system work when it can't be mapped to our data in a consistent way?" After observing the vendor's evasive reply, Karen went straight to her point, "How can a metadata system work properly when the main data system isn't configured consistently?" After doing some research on her own, the data steward learned what she had suspected all along: If you don't have a clear sense of the data in your system, you can't expect a metadata system to help you better use and manage the data.
Karen then gave the vendor a choice: take the system's data quality issues seriously, or have the contract canceled so the RFP could be revised to include system analysis prior to metadata implementation.