Data Security: 8 Privacy Principles Every IT Pro Should Know

Compliance / Privacy
  • Smaller Small Medium Big Bigger
  • Default Helvetica Segoe Georgia Times

Securing your data involves understanding data privacy regulations and the principles behind those rules

by Graham Williamson

Editor’s note: This chapter is excerpted from chapter 11 of  Identity Management: A Business Perspective (MC Press, 2017).

Identity management has a direct impact on the security of identity data and regulatory controls over the collection, storage, and usage of identity information.

As we become more connected, organizations, both commercial and public sector, will increasingly store more data about us. This will be driven by the desire of companies to understand us so that they can sell us more products and services, and by governments that want to leverage their identity information to adopt a “digital first” strategy for providing services. As these organizations extend their data repositories, the general public expects this data to be adequately managed to protect their privacy. However, members of the public are becoming increasingly scared because they know systems are vulnerable and they doubt the capability of organizations, particularly government, to adequately protect it.

Organizations can therefore expect more, not less, regulation. Rather than resist this, we need to embrace it, plan for it, and deploy systems that are compliant. As with most regulatory issues, this needs to be done this before it has to be done.

Regulation Summary

The major regulations that affect data security are as follows:


The payment card industry (PCI) provides guidelines on the correct treatment of information such as credit card details and provides an auditable standard. It is most important that, as e-commerce continues to grow, organizations safeguard their credit card information handling to avoid revenue loss from repudiated transactions and violation penalties. Payment gateways must adhere to PCI guidelines and submit to periodic audits.


The purpose of the US Health Insurance Portability and Accountability Act (HIPAA) is to protect health information. While the main focus of health information protection has been embodied in personal health record controls, HIPAA goes beyond that to set regulation for security controls that manage access to protected health information. It stipulates how hospitals and health centers should ensure protection of sensitive data.


The Sarbanes-Oxley Act (SOX) in the US does not directly impinge on data privacy. It does seek to regulate, and provide useful guidelines for, corporate governance, including auditor independence and the need to disclose data loss that has detrimental effects.


The Family Education Rights and Privacy Act (FERPA) seeks to manage public schools and colleges in the US and provides a framework for any educational jurisdiction to follow in the collection and management of student education records.


The Gramm-Leach Bliley Act (GLBA) forms part of privacy regulation in the US and provides guidelines for the control of personal information, particularly in the financial industry. Banks, insurance, and wealth management companies are covered by the GLBA, which requires them to implement safeguards in regard to client personal information.

Privacy and Data Protection

Some countries have focused legislation specifically on privacy of sensitive data, while others have addressed privacy in the wider context of data protection.


The concept of privacy is quite broad. We all think we know what it means, but there are many viewpoints as to how best to define it. Although there are dictionary definitions of what it is, it is more meaningful to define what it means:

Privacy is the desire of individuals to control, or influence, the collection, usage, and disclosure of data about themselves.

The following terms are related to privacy, and need definition in their own right:

  • Confidentiality—seeks to control the release of identity data
  • Secrecy—is the intentional concealment of identity data
  • Anonymity—is the ability to conceal identity data

Privacy is largely a Western construct, and even among the developed countries there is wide divergence on the desire for privacy and toleration of its violation. In many Western European countries, particularly those that have experienced occupation during wartime, the desire for privacy is tempered by the desire for efficiency of government services. In other countries such as the US and the UK, there is a skepticism of government and its ability to keep identity information private. In these geographies, privacy is highly regarded, and efficiency of government services will typically be subjugated to privacy protection.


Most privacy legislation has at its core certain principles that, if followed, will keep user information private and avoid contravention of privacy legislation, with its attendant costs. Table 1 lists these principles.

Table 1: Privacy principles




Organizations that collect personally identifiable information (PII) should publish a policy that states how the organization collects, stores, and uses such information.

Instruction must be provided regarding how an individual may gain access to the organization’s information to validate its accuracy. Instruction should be provided regarding how an individual may lodge a complaint about the information that is kept on them and how the organization will deal with such complaints, including expected timeframes.

Collection of data

Organizations may only collect PII data that is required for a specific transaction. They can’t collect data that they “might need in the future” or collect extra data for marketing purposes unless they expressly state that and seek the subject’s consent. This means that, if an organization wants to use transaction data for big data analysis for marketing purposes, they must de-identify it.

A common violation of privacy policy is committed by software testers when they use a client database for test purposes, often in the cloud. In such an instance, the identity data should be obfuscated by randomizing first and last names, address detail, and phone numbers.

When collecting data, the organization must ensure that informed user consent is collected; that is, advice as to the stated use of the information must be provided.

If a transaction does not need the explicit identification of a user, there must be the option of completing a transaction anonymously or with a pseudonym.

Disclosure of personal information

It is forbidden for organizations to disclose PII to other organizations, or even other divisions within the same organization, unless expressly agreed to by the subject. This means that if a bank collects PII for the purpose of opening a bank account, they cannot provide this information to their insurance branch without the subject’s express permission.

Cross-border disclosure

Organizations that store PII information on citizens of the country in which they undertake business must not store the information in a jurisdiction that does not have legislation at least as stringent as in the home country. For instance, if a call center is established in a low-cost jurisdiction that has weaker privacy legislation, the organization must deploy specific controls to ensure the privacy of persons in the data file used by call center representatives.

Use of government identifiers

There are constraints on the use of government identifiers as an index on data files. For instance, a department of motor vehicles might use a driver’s license number as an index, but another department should not use driver’s license numbers as an index for their data stores. Equally, commercial organizations that collect driver’s license numbers (e.g., car rental companies) should create their own index fields and not use driver’s license numbers for this purpose. This does not preclude the use of a driver’s license number to locate a user record.

Quality of PII data

Organizations that collect PII data must take steps to maintain its currency. Once a data record has passed its date of currency, there are two options: the company must engage resources to validate its accuracy and correct it if it’s wrong, or it must delete the data record. The date of refresh should be stated in the organization’s privacy policy. However, it’s generally agreed that address data is sufficiently out of date after three years.

Security of PII data

Organizations must deploy appropriate security measures, commensurate with the sensitivity of the data in question. Data must be protected from misuse, interference, unauthorized access, modification, and disclosure. This includes data backups and files used for testing purposes.

A related control is the necessity to delete or destroy PII information that is no longer needed; it’s not permissible to maintain a data file containing PII data just because it might be of use in the future.

Access and correction of PII data

If an entity holds personal information about an individual, the entity must, on request by the individual, give them access to the information. This means that processes must be established to facilitate access, and correction processes should be published in the organization’s policy statement with an indication of the timeframe that the subject should expect for a response. If the subject of a record containing PII data requests correction of a data element, it must either be changed or annotated.

While an initial reaction might be that these requirements are onerous, they do in fact make good sense. Organizations are well advised to adopt these principles and avoid the cost and embarrassment of data loss that might otherwise occur.