Data integrity vs. data quality: Is there a difference?

Briefly, sure. Once we discuss information integrity, we’re referring to the overarching completeness, accuracy, consistency, accessibility, and safety of a corporation’s information. Collectively, these components decide the reliability of the group’s information. Information high quality makes use of these standards to measure the extent of information integrity and, in flip, its reliability and applicability for its meant use. Information high quality and integrity are very important to a data-driven group that employs analytics for enterprise selections, provides self-service information entry for inside stakeholders and supplies information choices to clients.

Information integrity

To realize a excessive degree of information integrity, a corporation implements processes, guidelines and requirements that govern how information is collected, saved, accessed, edited and used. These processes, guidelines and requirements work in tandem to:

• Validate information and enter
• Take away duplicate information
• Present information backups and guarantee enterprise continuity
• Safeguard information through entry controls
• Preserve an audit path for accountability and compliance

A corporation can use any variety of instruments and personal or public cloud environments all through the information lifecycle to keep up information integrity by means of one thing generally known as data governance. That is the apply of making, updating and constantly imposing the processes, guidelines and requirements that stop errors, information loss, information corruption, mishandling of delicate or regulated information, and information breaches.

The advantages of information integrity

A corporation with a excessive degree of information integrity can:

• Improve the probability and velocity of information recoverability within the occasion of a breach or unplanned downtime
• Defend towards unauthorized entry and information modification
• Obtain and preserve compliance extra successfully

Good information integrity also can enhance enterprise determination outcomes by rising the accuracy of a corporation’s analytics. The extra full, correct and constant a dataset is, the extra knowledgeable enterprise intelligence and enterprise processes turn out to be. Consequently, leaders are higher geared up to set and obtain objectives that profit their group and drive worker and client confidence.

Information science duties resembling machine studying additionally enormously profit from good information integrity. When an underlying machine studying mannequin is being educated on information data which might be reliable and correct, the higher that mannequin can be at making enterprise predictions or automating duties.

The several types of information integrity

There are two fundamental classes of information integrity: Bodily information integrity and logical information integrity.

Bodily information integrity is the safety of information wholeness (which means the information isn’t lacking necessary info), accessibility and accuracy whereas information is saved or in transit. Pure disasters, energy outages, human error and cyberattacks pose dangers to the bodily integrity of information.

Logical information integrity refers back to the safety of information consistency and completeness whereas it’s being accessed by completely different stakeholders and purposes throughout departments, disciplines, and places. Logical information integrity is achieved by:

• Stopping duplication (entity integrity)
• Dictating how information is saved and used (referential integrity)
• Preserving information in an appropriate format (area integrity)
• Guaranteeing information meets a corporation’s distinctive or industry-specific wants (user-defined integrity)

How information integrity differs from information safety

Information safety is a subcomponent of information integrity and refers back to the measures taken to forestall unauthorized information entry or manipulation. Efficient information safety protocols and instruments contribute to sturdy information integrity. In different phrases, information safety is the means whereas information integrity is the aim. Information recoverability — within the occasion of a breach, assault, energy outage or service interruption — falls beneath the realm of information safety.

The implications of poor information integrity

Human errors, switch errors, malicious acts, inadequate safety and {hardware} malfunctions all contribute to “unhealthy information,” which negatively impacts a corporation’s information integrity. A corporation contending with a number of of those points dangers experiencing:

Poor information high quality

Low-quality information results in poor decision-making due to inaccurate and uninformed analytics. Diminished information high quality may end up in productiveness losses, income decline and reputational harm.

Inadequate information safety

Information that isn’t correctly secured is at an elevated danger of a knowledge breach or being misplaced to a pure catastrophe or different unplanned occasion. And with out correct perception and management over information safety, a corporation can extra simply fall out of compliance with native, regional, and world rules, such because the European Union’s Normal Information Safety Regulation.

Information high quality

Information high quality is actually the measure of information integrity. A dataset’s accuracy, completeness, consistency, validity, uniqueness, and timeliness are the information high quality measures organizations make use of to find out the information’s usefulness and effectiveness for a given enterprise use case.

How one can decide information high quality

Information high quality analysts will assess a dataset utilizing dimensions listed above and assign an total rating. When information ranks excessive throughout each dimension, it’s thought of high-quality information that’s dependable and reliable for the meant use case or software. To measure and preserve high-quality information, organizations use information high quality guidelines, often known as information validation guidelines, to make sure datasets meet standards as outlined by the group.

The advantages of excellent information high quality

Improved effectivity

Enterprise customers and information scientists don’t need to waste time finding or formatting information throughout disparate techniques. As a substitute, they’ll readily entry and analyze datasets with larger confidence. Further time is saved that may have in any other case been wasted on appearing on incomplete or inaccurate information.

Elevated information worth

As a result of information is formatted constantly and contextualized for the consumer or software, organizations can derive worth from information that will have in any other case been discarded or ignored.

Improved collaboration and higher decision-making

Excessive-quality information eliminates incongruency throughout techniques and departments and ensures constant information throughout processes and procedures. Collaboration and decision-making amongst stakeholders are improved as a result of all of them depend on the identical information.

Diminished prices and improved regulatory compliance

Excessive-quality information is simple to find and entry. As a result of there isn’t any have to re-create or monitor down datasets, labor prices are decreased, and guide information entry errors turn out to be much less probably. And since high-quality information is simple to retailer within the right setting in addition to accumulate and compile in necessary experiences, a corporation can higher guarantee compliance and keep away from regulatory penalties.

Improved worker and buyer experiences

Excessive-quality information supplies extra correct, in-depth insights a corporation can use to offer a extra customized and impactful expertise for workers and clients.

The six dimensions of information high quality

To find out information high quality and assign an total rating, analysts consider a dataset utilizing these six dimensions, often known as information traits:

1. Accuracy: Is the information provably right and does it replicate real-world information?
2. Completeness: Does the information comprise all related and out there info? Are there lacking information parts or clean fields?
3. Consistency: Do corresponding information values match throughout places and environments?
4. Validity: Is information being collected within the right format for its meant use?
5. Uniqueness: Is information duplicated or overlapping with different information?
6. Timeliness: Is information updated and available when wanted?

The upper a dataset scores in every of those dimensions, the larger its total rating. A excessive total rating signifies {that a} dataset is dependable, simply accessible, and related.

How one can enhance information high quality

Some frequent strategies and initiatives organizations use to enhance information high quality embody:

Information profiling

Information profiling, often known as information high quality evaluation, is the method of auditing a corporation’s information in its present state. That is accomplished to uncover errors, inaccuracies, gaps, inconsistent information, duplications, and accessibility obstacles. Any variety of information high quality instruments can be utilized to profile datasets and detect information anomalies that want correction.

Information cleaning

Information cleaning is the method of remediating the information high quality points and inconsistencies found throughout information profiling. This consists of the deduplication of datasets, in order that a number of information entries don’t unintentionally exist in a number of places.

Information standardization

That is the method of conforming disparate information property and unstructured massive information right into a constant format that ensures information is full and prepared to be used, no matter information supply. To standardize information, enterprise guidelines are utilized to make sure datasets conform to a corporation’s requirements and desires.

Geocoding

Geocoding is the method of including location metadata to a corporation’s datasets. By tagging information with geographical coordinates to trace the place it originated from, the place it has been and the place it resides, a corporation can guarantee nationwide and world geographic information requirements are being met. For instance, geographic metadata might help a corporation be certain that its administration of buyer information stays compliant with GDPR.

Matching or linking

That is the tactic of figuring out, merging, and resolving duplicate or redundant information.

Information high quality monitoring

Sustaining good information high quality requires steady information high quality administration. Information high quality monitoring is the apply of revisiting beforehand scored datasets and reevaluating them primarily based on the six dimensions of information high quality. Many information analysts use a knowledge high quality dashboard to visualise and monitor information high quality KPIs.

Batch and real-time validation

That is the deployment of information validation guidelines throughout all purposes and information sorts at scale to make sure all datasets adhere to particular requirements. This may be accomplished periodically as a batch course of, or constantly in actual time by means of processes like change information seize.

Grasp information administration

Grasp information administration (MDM) is the act of making and sustaining an organization-wide centralized information registry the place all information is cataloged and tracked. This provides the group a single location to rapidly view and assess its datasets no matter the place that information resides or its sort. For instance, buyer information, provide chain info and advertising and marketing information would all reside in an MDM setting.

Information integrity, information high quality and IBM

IBM provides a variety of built-in information high quality and governance capabilities together with information profiling, information cleaning, information monitoring, information matching and information enrichment to make sure information shoppers have entry to trusted, high-quality information. IBM’s data governance solution helps organizations set up an automatic, metadata-driven basis that assigns information high quality scores to property and improves curation through out-of-the-box automation guidelines to simplify information high quality administration.

With information observability capabilities, IBM might help organizations detect and resolve points inside information pipelines sooner. The partnership with Manta for automated information lineage capabilities allows IBM to assist shoppers discover, monitor and forestall points nearer to the supply.

Learn more about designing the right data architecture to elevate your data quality here.