It is believed that a person browses the web at an average rate of 20 megabytes per hour, an office of 100 would take about 142,693 years to consume the data that was generated daily four years ago. This data takes various forms, including social media posts, emails, tracking information from cloud services, and teleconference chats. Without a strategy to organize and manage this data, it would be difficult to effectively utilize it, let alone make sense of it.
Data governance is the system that manages the sourcing, availability, usability, security, and disposal of data in an industrial or enterprise setting. Proper data governance is not only needed to make sure that the data utilized for business decisions is consistent and meets the needs of the business, but that its use is secure and coheres to fiduciary and legal responsibilities.
The policies behind a company’s data governance is typically formed by a governance team, a committee of stakeholders that steer policy toward what is best for the company, and data stewards that are in charge of enforcing the policy. Along with other data management teams, data governance helps to form the strategy that determines a company’s disposition toward IT.
Capella Covers: Data Governance
- What is and isn’t Data Governance?
- Data Governance vs. Data Management
- Pros and Cons of Data Governance
- Data Governance Processes
What Is Data Governance?
Data is simply just bits of information. Depending on the system they were collected on, they may take different formats and forms, even within the same company. This may include—just for a typical sales department—orders from customers, credit card transactions from banks, receipts from the sales software, inventory information from the warehouse, and vendor invoices. There may also be time adherence information from the sales employees, logistical data from the showrooms, security and surveillance data, and many other data feeds. Data governance is simply the strategy a company uses in dealing with this information.
Data governance can consist of properly formatting data in a way that it is usable by all systems in a company, restricting access to sensitive data, establishing proper containment to make sure that data does not leak from the company, and developing strategies to identify and retire data that is no longer relevant. Much of the data that a company may receive may be transitional, where there is an expectation by the originator that it will be held in confidence.
Violating this confidence would erode the trust held for the company. Other data may be protected by federal or state data protection laws, like HIPAA or the Data Protection Act. Violating these laws would expose the company to legal liabilities or to excessive fines. Finally, some data may be “noisy,” with too much extraneous information attached to make it useful. Policies may be needed to determine what can be stripped and what should be saved.
Data Governance In The Real World
A recent example of why this is important is Twitter. Like most social media companies, Twitter made its money by selling access to its user contents to marketers, data analysis firms, and third-party app makers.
For much of its history, Twitter did not take definitive steps to safeguard this data, despite its users expecting that their tweets would be safe. During the 2016 US Presidential Elections, this openness was used, per Twitter, to help influence voters via the use of fake accounts. The revelation of this deeply undercut confidence in the company. While the company has since enhanced its data stewardship practices, data breaches remain an issue, with a data breach affecting 5.4 million users being detected in August 2022.
A data governance plan is a dynamic, living strategy. It changes to meet the needs of the company and the times. While there may be nothing that will keep your company’s data perfect and perfectly safe—in 2021, there were 1,862 data breaches—a strong plan will mitigate a company’s risk while ensuring that its data is the most useful.
What Is NOT Considered Data Governance?
Data governance is only a part of IT. It cannot replace proper data management or information infrastructure. Below we break down the difference between data governance and data and master data management
Data Governance Is Not The Same As Data Management
For example, data governance is not data management. Data governance is a subset of data management, which specifically deals with the establishment of policies and standards to maintain and ensure data integrity and protection. Data governance sets the policies that will define a company’s data management, including rules for data management, pipelines for data from one system to another, ETL (Extract, Transform, Load) policies that format data from one system to a company standard, data warehousing rules, and needs for a company’s data architecture.
Data Governance Is Not Master Data Management
More specifically, data governance is different from Master Data Management (MDM). Master data is the formatted, consistent set of information a company uses to operate, including data on customers, leads, suppliers, sites, and its chart of accounts. MDM is the methodology that a company uses technology to help manage master data and makes it accountable.
Like data management, MDM utilizes principles of data governance. MDM deals with the creation of a “single version of the truth” (SVOT), where differing versions of a firm’s master data is identified and rectified to the overriding corporate standard. MDM is essential to eliminate inefficiencies that may arise from working from multiple versions of the “truth.” MDM usually involves data verification, ETL, and data reconciliation on some level.
Pros And Cons Of Data Governance
Like any other business application, data governance has pros and cons. In determining a data governance structure for your business, you must consider if it will be a fit for your business needs.
Pros Of Data Governance
- Accurate, Reliable Data - For businesses, having a trustworthy data pipeline is essential. Without “safe” data, a company cannot make accurate business determinations. A business cannot be expected to make timely decisions with multiple, differing data streams, as the business will have to determine “truthfulness” and relevance on a case-by-case basis.
- A “Single Version of the Truth” - Different systems may interpret and format data differently, resulting in differing interpretations of data. Additionally, older versions of datasets may create a varying profile of the “truth.” A strong data governance would include rules for ETL, data reconciliation, and data retirement, meaning that the data available for business decisions can safely be considered master data.
- Legal and Regulatory Compliance - There are data types that must be protected, such as personally identifiable information (PII), financial account information, health information, and contractually secured private information. Failure to secure this information may expose the company to grave legal and reputational consequences. Despite this, information may be needed by other systems. A proper data governance plan would dictate what data must be redacted, what data cannot be shared, and what data can be shared internally and externally.
- Cost Savings - Multiple versions of the “truth” would require multiple infrastructures to support them. By maintaining a SVOT, redundancies can be removed, creating a cost reduction. Audits would also be quick and easy and day-to-day operations would be simpler with a single truth. Waste will be reduced from the elimination of incorrect or outdated information and customer service would be improved through the availability of accurate data.
Cons Of Data Governance
- Need for Additional Resources - It will take more staff and resources to manage a data governance plan. This may include the addition of data stewards, an oversight committee, and additional data storage, such as siloed data storage. For small organizations, it may be unreasonable to have a full data governance plan.
- Siloed Data - Sensitive data may need to be separated for various reasons. This may lead to a lack of needed system inter-connectivity regarding data transfer. This may be fueled by additional data sources, the natural growth of new technologies, and evolving infrastructures. There may need to be a special effort in place to avoid siloed data under a data governance system.
- Need for Dedicated Leadership - Not everyone is data literate. A company’s chief IT officer may not know enough about the intricacies of data governance to make proper decisions. It may be needed to have dedicated, informed leadership knowledgeable in data governance to coordinate your company’s plan. Additionally, you may need to hire and train data stewards to implement this plan and make sure members of the oversight committee stay knowledgeable about requirements such as compliance with GDPR and CCPA.
Data Governance Processes
Data governance processes typically cover five components: data architecture, data quality, data management, data security, and data compliance. Plans and policies regarding a company’s data governance must address all five of these components and must be cognizant of the needs of these specific areas.
The purpose of a competent data governance plan is to make sure that data stakeholders are involved in the management and integrity of the data flow. This includes proactive policies, such as establishing a secure protocol for data-gathering web pages and making sure that current security measures meet needed standards. This also includes reactionary measures, such as the prompt response to a data breach and the recognition of deficiencies in security and integrity measures. Depending on the nature of the company’s data governance process, this can require a committee’s consultation, based on the needs of the company. Such decision-making can also be made by a Chief Data Officer or by a similar executive in charge of responding to data governance needs.
Data Governance Tools And Technology
The proper tools are needed to manage a proper data governance plan. This would include a centralized database for the storage of the data, tools to catalog the data, security tools such as firewalls and antivirus checkers, collaborative tools that allow different teams to access data in compliance to your company’s rules, and tools that help publicize and share data governance policies.
Many companies offer solutions as software (SaaS) packages that simplify and unify many of these tools, making it easier for a company to establish a comprehensive data governance policy quickly.
Why Your Company Needs A Data Governance Plan
It may be that you have never considered the need for data integrity or for master data. But, if your company deals with multiple data sources, such as customers, vendors, or employees, at some point data congestion and data integrity will become an issue. While the image of the small business person shuffling through receipts may be quaint, it is also inefficient. Such inefficiencies will cost the business money and confusion as it grows.
A strong data governance plan—which details the company’s strategies for data gathering, filtering, cohesion, security, storage, and disposal—will help safeguard the company’s data and integrity. It is essential to meeting governmental and industry data safeguarding regulations, to maintain the “single version of the truth,” and to avoid costly redundancies and duplicates. A comprehensive data governance plan may be the best security plan against this ever-changing, increasingly hostile digital world.
For more information on how to adapt to a changing digital world, check out Capella’s solutions and services that help you find the end-to-end platform for your data quality, data governance, and master data management needs. Our key advantages provide:
- Open, fully extensible platform
- Ultrafast search experience
- Simplified management capabilities
- Ability to integrate any data source
- Enterprise-grade security
To get started, send us a message today and let us know how we can help your business transform complex data problems.
Is a solution and ROI-driven CTO, consultant, and system integrator with experience in deploying data integrations, Data Hubs, Master Data Management, Data Quality, and Data Warehousing solutions. He has a passion for solving complex data problems. His career experience showcases his drive to deliver software and timely solutions for business needs.