Skip to main content

Data Quality

We place a premium on Data Quality as this is essential to build trust and confidence in your organisation's Data. Data stewards are expected to understand their Data and be able to define the rules that should be implemented to control the quality.

Monitored continuously, resolved at source​

Data quality is a live feed rather than a periodic audit. Issues are tracked as data arrives, and records that do not conform to the rules defined for them are rejected back to the source system rather than corrected downstream — so the correction happens once, where the data is created, instead of in every report that reads it.

That only works if the business functions producing the data own its quality, which is why measurement is published by function rather than held centrally. Accountability is the mechanism; the dashboards are how it is made visible.

For the dashboards themselves, see the quality guides: the function dashboard and steward dashboard, the parking lot holding rejected records, and how the score is calculated.

The six DAMA dimensions​

Data Quality is measured in line with the six dimensions from the DAMA Data Management Body of Knowledge, which are:

Accuracy​

Data that attempts to model real-world objects or events.

Example:

  • Incorrect spelling of a person's name
  • Incorrect address for delivery
  • Company ABN lookup check against ASIC records

Completeness​

All required records and values should be available with no missing information. With completeness, the stored data is compared with the goal of being 100% complete.

Example:

  • An address on a membership form. If three forms out of 100 are missing addresses, the data, regarding addresses, is 97% complete.

Consistency​

This dimension is about a lack of difference when two or more data items are being compared. Items of data taken from multiple sources should not (in an ideal world) conflict with one another.

Example:

  • A school’s database having a student’s date of birth showing the same format and value in both the school register and the documents sent from the school the student is transferring from.

Timeliness​

The data’s actual arrival time is measured against the predicted, or desired, arrival time.

Example:

  • A take away promises to deliver your food in 40 minutes, however the person taking the order only enters the Data into the system after the deliver was due, thus the KPI is not met.

Uniqueness​

Data is properly identified and only recorded once. When data is unique, no record exists more than once within a database. Each record can be uniquely identified, with no redundant storage.

Example:

  • A person should not appear more than once in a database becuase of a shortened version of their name. Thomas or Tom could be duplicate records of the same person.

Validity​

Data closeness to pre-defined business rules or a calculation. When these rules are applied, the data falls within defined parameters or conforms to the syntax of its definition. Examples

  • Based on Business Rules or Calculation
  • Based on Validity for range of values
  • Sequencing Order.

Example:

  • Invoices order, receiving Invoice 1000 before Invoice 999 breaks the expected sequence.
  • Personal Identity check using forms or Identification on file.
  • Credit Card Number check - right number of Digits.