You’ve heard the phrase “garbage in, garbage out.” Bad data causes all sorts of trouble: poor decisions, angry customers, higher costs. Use this exercise to assess whether there are problems with your data.
- Gather. Assemble the last 100 data records your group used or created. For example, if your group takes customer orders, assemble the last 100 orders; if you create engineering drawings, assemble the last 100 drawings. Then focus on 10–15 critical data elements or attributes in the data record.
- Review. Ask two or three people with knowledge of the data to join you for a two-hour meeting. Working record by record, mark obvious errors, like a misspelled customer name or information that’s been placed in the wrong column. You may need to stop and discuss whether an item is truly incorrect, but try not to spend more than 30 seconds on a record.
- Summarize the results. Total the number of perfect records, and then determine what percentage of your data is accurate. If you have a data quality problem, target your efforts to fix it.
Adapted from "Assess Whether You Have a Data Quality Problem," by Thomas C. Redman
