Part 3: This is the third of eight posts on how to measure data quality. This post describes why accuracy is a good measure and how it’s used.
Accuracy states that a piece of information is what it should be. In our definition, a telephone is a telephone if it has the right number of valid numbers and the right format for the right country. For example, this is a valid number for the Netherlands +31104065777, however if the field is supposed to contain mobile numbers then it would be invalid.
Let’s say there are 70,000 records and only 65,000 are present (then we have a completeness of 92.8%, see part 2 post on completeness.) If only 60,000 phone numbers are valid numbers then accuracy is 92.3% of all phone numbers present, but only 85.7% accurate for the whole database.
Accuracy is a percentage measurement of the accuracy of the data that is actually present.
To improve accuracy is a clear benefit for all. Having more accurate phone numbers allows better telemarketing, telesales or teleprospecting campaigns. Having more accurate addresses reduces costs through returned email or returned shipments, increases sales through more prospects being reached and increases the ROI on your marketing campaigns. Having more accurate emails can also increase email campaign results.
Having good data entry standards, batch software to improve degraded data and importing only accurate data will ensure accuracy is maintained.
The other dimensions are:
- Completeness (Part 2 of 8)
- Consistency (Part 4 of 8)
- Conformity (Part 5 of 8)
- Currency (Part 6 of 8)
- Duplication (Part 7 of 8)
- Integrity (Part 8 of 8)