Try Data Enso!
  • FR

Data Quality - Data KPIs

Make sure your decisions are profitable!

Try Data Enso Solutions
Contact us

Data KPIs

Meeting a company’s goals depends heavily on the quality of the data collected, stored and processed. The better the quality, the more likely the company will optimize its time and make sound and profitable strategic decisions.

Companies that understand the importance of data quality and have implemented continuous improvement processes have seen a 15-20% increase in revenue due to “intelligent” decision making. Conversely, US companies lost an estimated $3 billion in 2016 due to decisions based on poor quality data.

Data quality improvement processes must integrate both monitoring and measurement tools, as well as solutions to optimize incoming data (collection format, normalization, integrity, security…) and specific processing methods for internal use.

Companies base their assessment of data quality on seven characteristics.

1 - Consistency
Data is recorded consistently across all data collection points and destination tables. Two different treatments of the same data give the same result; the data must match or at least align.

Monitoring indicator: data range, variance calculation, standard deviation
2 - Accuracy
Data precision is determined by its accuracy. The precision rate is the difference between the measured value and the exact value of the data. The data shouldn't contain any error, be obsolete or be duplicated.

Monitoring indicator: error rate calculation
3 - Completeness
A high quality dataset is made up of complete data sets with sufficient usable information. High quality data has a positive impact on the decision making process.

Tracking indicator: percentage of data containing all necessary information
4 - Auditable
All data is stored and freely accessible. All processing and modifying actions on these data are recorded and logged: during an audit, who intervened, which action, the data is recorded and exportable.

Monitoring indicator: Rate of altered data, Rate of data not found
5 - Validity
The data is stored in the same format regardless of the point of collection, making it easier to use the data later on. Unique formatting rules should be established to ensure a high validity rate.

Tracking indicator: Percentage of data where all values are in the required format
6 - Uniqueness
Data is recorded only once in a single location. Duplicate entries should be avoided as they can lead to errors. Monitoring this indicator allows you to identify and avoid double data entry.

Monitoring indicator: Number or percentage of repeated values
7 - Obsolescence
The interval between the collection and the expected evaluations meets the company's needs. Data collection is carried out at the right time for the expected use. This makes it possible to monitor changes effectively.

Monitoring indicator: time variance
Previous slide
Next slide

Maintaining and improving the quality of your data is part of a continuous process within the company. Implementing Data Quality Management solutions impacts the quality level throughout the data’s life cycle. You need to choose the right DQM solutions for your needs and your expected quality objectives.

Data Enso has implemented additional metrics to complement these key data quality metrics and provide additional insight into the use of our DQM solutions and the availability of our services.

All of these activity monitoring criteria will be accessible in your My Data Enso space when our partnership is set up:

  • Number of data submitted for validation
  • Response rate of our solutions
  • Data validation rate (status "ok")
  • Company enrichment rate
  • Average response time
  • Service availability: server, website, API solutions

The essentials of data quality

Contact

+33 (0)1 47 72 79 61

Book an appointment

Newsletters