News & Insights

Implementing Privacy by Design

Zasio, Your Complete Information Governance Partner, in white over a purple-tinted city skyline

Privacy by Design is a framework that requires the proactive consideration of privacy in the design and operation of business practices and IT systems. While this might sound simple in concept, it can be difficult to implement in practice, especially if privacy is not your day-to-day role!

In this webinar, we simplify PbD with an overview of the concept and discussion of some of the key privacy principles and implementation techniques. We also provide you with some key elements that you can use in your office to help your PbD framework move from concept to good practice. Whether you’re a privacy newbie or a daily practitioner, there is something in this webinar for you!

Key takeaways: Implementing Privacy by Design

  • Privacy by Design means building privacy in from the beginning, before a breach or misuse, rather than reacting after an event.
  • The less personal data you hold and the fewer people who can access it, the less damaging a breach will be, so collect only what you truly need.
  • Start embedding privacy with a privacy impact assessment to understand the scope, risks, and measures needed to mitigate the company's exposure.
  • Training on personal data should not stop at onboarding; keep it ongoing so privacy stays top of mind throughout employment.
  • Before drafting a privacy notice, understand the data lifecycle: how it is collected and used, who it is shared with, how long it is kept, and disposal.
  • Give notice in plain, clear language at or before the time of collection, then collect consent and keep a log showing that consent was gathered.

Browse all Zasio webinars

Audio-described version
Transcript: Implementing Privacy by Design

Duration: 0:21:59

Speakers: Soo Kang, General Counsel and Director of the Consulting Division, Zasio Enterprises

Lightly edited machine transcript. Slides are summarized where they carry information not spoken aloud.

Soo Kang: Thank you, everyone, for joining today's webinar on Privacy by Design. My name is Soo Kang, and I am the General Counsel and Director of the Consulting Division at Zasio Enterprises. Before we get started, a quick housekeeping matter: the obligatory attorney disclaimer that this presentation is conducted for informational purposes only, and not for the purposes of providing legal advice. With that taken care of, let's move on to the presentation.

What are we going to cover today? We will start by laying down the foundation for why Privacy by Design is needed. Then we will cover the seven principles behind the Privacy by Design framework. After that, we will move to a high-level overview of implementation stages, and finish with a sample scenario.

Let's start with the why. Why is Privacy by Design a framework that needs to be implemented or considered by companies?

[Slide: Quote attributed to Bruce Schneier, American Cryptographer, Privacy Specialist, and Writer (Schneier on Security, March 4, 2015).]

In tackling the topic of the worth of personal data and the line between privacy and convenience, Bruce Schneier compares data to pollution and privacy as an environmental challenge. In other words, there is a lot of information about people out there that is collected, produced, and used. The challenge is, how do you effectively manage it?

[Slide: "Data brokers have extensive profile information on entire populations": examples of data on consumers provided by Acxiom and Oracle; Acxiom provides up to 3,000 attributes and scores on 700 million people in the US, Europe, and other regions.]

To give you a sense of the enormity of this problem, there was an article from Fast Company in March of this year about the recently passed Vermont law requiring data brokers to register with the Secretary of State. Within that article, they discussed how much information can be collected from individuals to create profiles. Citing to this visual, they noted that in 2017, data giant Acxiom provided up to 3,000 attributes about individuals on 700 million people. In 2018, the number was 10,000 on 2.5 billion consumers. While the growth within that one year is unbelievable, you can see that even in 2017, Oracle provided more than 30,000 attributes on 2 billion consumer profiles.

And look at the categories of information that they are collecting. It's about people's hobbies, purchases, search histories, political views, religion, financial stability, household data, and so on. While this is not surprising, as we are living in a connected world and our activities can be tracked through our phones, computers, watches, and other connected devices, even with that understanding, it's still daunting, the sheer amount of information being collected about us as individuals.

[Slide: 92% agree companies must be proactive about data protection. 69% of consumers believe companies are vulnerable to hacks and cyberattacks; 25% of consumers believe most companies handle their sensitive personal data responsibly; 15% think companies will use that data to improve their lives; 10% of consumers feel they have complete control over their personal information. Source: Consumer Intelligence Series: Protect.me, PwC, Dec. 2017.]

It's no wonder that this has led to trust issues with regard to how companies manage, use and secure the personal data that they collect. 69% believe that there is inadequate security employed by companies. Only 25% believe companies are responsible in handling their sensitive personal data. Only 15% believe that the data collected is actually being used by companies to improve their lives, with only 10% feeling that they have complete control over their personal information. That said, one thing that everyone can agree on is that companies need to be more proactive about data protection, and this is where Privacy by Design comes into play.

The Privacy by Design framework was initially developed by Dr. Cavoukian while she was serving as the Information and Privacy Commissioner of Ontario, Canada. Its goal is to proactively embed privacy into the design of technologies, networks, and business practices.

[Slide: Quote from Tim Cook, Apple CEO (source dated February 13, 2015).]

As Tim Cook stated during the White House Cybersecurity Summit in February of 2015, people have entrusted companies with their personal data, and in return there is an obligation to protect. This quote is reflective of an attitude shift regarding ownership and control of personal data, especially in the United States: that of imposing greater accountability on companies for the collection, use, and security of personal data. To achieve this, companies must think about privacy impacts from the outset, which aligns with Privacy by Design. So let's dive into the seven foundational principles in the Privacy by Design framework.

[Slide: Principle 1: Proactive not reactive, preventative not remedial. PREPARE: build privacy in from the beginning, before a data breach.]

This first principle is about preparation, and it's the overarching principle behind the framework. The last place you want to be as an organization, especially in today's legal and regulatory environment, is to be in a position to second-guess yourself about whether you should or should not have done something to avoid a disastrous event, be it a breach or misuse of personal data. You need to understand the consequences of decisions being made, and that even small concessions may lead to big and unwarranted outcomes.

[Slide: Principle 2: Privacy as a Default Setting. BASELINE: Provide maximum privacy protection as a baseline.]

Privacy as a default setting, the second principle, is about lowering the data security risk as a baseline. Essentially, the less data you have, the less you have allowed access to, and the less damaging a breach will be. Even without the data subjects doing anything, personal data should be automatically protected. This can be done by following the universal privacy practices of purposeful collection; lawfulness, fairness, and transparency; and minimalization. Purposeful collection means: why are you collecting it, and do you really need it?

Lawfulness, fairness, and transparency is the relationship you have with the data subjects, providing that adequate notice and being able to obtain a consent, informed consent. And minimalization means minimizing access to, limiting the use of the personal data to the purpose, and securely disposing when no longer necessary for that purpose or to satisfy legal requirements.

[Slide: Principle 3: Privacy Embedded into Design. EMBED: embed privacy when developing technologies and business practices.]

The third principle, of embedding privacy, means that privacy should not be an afterthought, but an essential core component of the functionality being delivered. For this, there are three key pillars to embedding privacy into design. Be holistic. Consider the broader context, not the narrow scope of the particular project. Be integrative.

Consider the interests of all stakeholders, including those of the data subject. And be creative. Consider alternatives, because current practices may not be acceptable. So how do you accomplish this? Well, you begin with a privacy impact assessment to understand the scope, risks, and measure needed to embed privacy to mitigate risks and exposures of the company.

[Slide: Principle 4: Positive Sum, Not Zero Sum. FULL FUNCTIONALITY: Accommodate all legitimate interests and objectives in a positive-sum "win-win" manner, not through a zero-sum approach where unnecessary trade-offs are made.]

This fourth principle, about full functionality, means that in developing a product or service, all legitimate interests should be accommodated. This is extremely difficult to accomplish, as practically, companies are more attuned to a risk-based approach with consideration of both risks and costs, rather than focusing on a win-win scenario. So in my mind and in my opinion, this is the most difficult thing to accomplish within the framework. But even if you are not able to reach a win-win approach or proceed in that fashion, just starting the conversation with this principle will lead to the benefits in the overall design process.

[Slide: Principle 5: End-to-End Security. PROTECTION: secure lifecycle protection, end-to-end.]

The fifth principle, about end-to-end security, means that privacy must be continuously protected throughout its lifecycle. Collaborate with stakeholders to identify appropriate security measures aligned to the sensitivity of the personal data that is being processed. Not all data are the same.

[Slide: Principle 6: Visibility and Transparency. TRUST: Visibility and transparency are essential to establishing accountability and trust.]

Visibility and transparency goes to what we discussed already. And that is, based on the stats we went over at the beginning, companies are already working behind the eight ball when it comes to trust, and it will take a lot of work to gain that trust back. To accomplish this, focus on accountability, openness, and compliance. Document, communicate, and assign responsibility. Be open about privacy practices with data subjects, and develop a way to monitor compliance, as well as allowing data subjects to redress errors that they find in the information about them.

[Slide: Principle 7: Respect for User Privacy.]

Lastly, keep the focus on the data subjects, and keep them top of mind throughout the project. Remember, data subjects have the greatest vested interest in the management of their personal data.

Now that we went over the framework of Privacy by Design, let's move to the implementation stages.

[Slide: Quote: "Good design begins with honesty, asks tough questions, comes from collaboration and from trusting your intuition." Freeman Thomas, automobile and industrial designer.]

This quote is not related to privacy, obviously, but it is on target for implementing Privacy by Design. Intuition does come a long way, so don't ignore the red flags. For example, in past presentations, I talked about the creepy factor line, as there is no gray when it comes to privacy, and even the best of intentions can lead to bad results.

A good example of this is a recent action brought by the FTC related to the so-called stalking apps. These mobile applications were designed to let employers and parents monitor the physical movements and online activities of their employees and children. We can see some practical application behind these apps and their benefits, but their design should have raised significant red flags regarding their potential misuse. On top of this, the developers did not impose restrictions on use to the stated purpose. It enabled the ability to hide the app by allowing an icon to disappear from the mobile screen, and provided instructions on bypassing device manufacturers' restrictions.

And compounding some of the issues that they faced was inadequate security on their end to safeguard the information that was collected. Not surprisingly, this led to both misuse of the application and breach of the cloud storage database, and ultimately, being on the radar of the FTC. Through a properly implemented Privacy by Design framework, a lot of these could have been identified and mitigated at the outset.

So how do you implement Privacy by Design? From a 10,000-foot perspective, the four main stages of implementing are purpose identification, design and development, data collection and use, and archive and disposal. We will discuss each in varying detail.

Identifying the purpose. For this step, you really need to understand what it is that you are trying to achieve and how you are going to achieve it, from resources tapped for support, both internal and external, to the types of data that would be collected. Be systematic in the approach, and develop standard questions to answer in this process.

Then moving on to design and development: conduct an early assessment to understand the privacy impacts of the project. This step is really about the visualization of the purpose identified. So the parties and data collected have been identified. Now how are you going to accomplish it? What does it look like in reality, and what hurdles will you face?

This provides the ability to forecast and be proactive in addressing those privacy-related issues and challenges. In conducting that assessment, you need to first identify key topical areas to cover, then come up with the lines of inquiry. As an example, this slide provides some topics and questions related to key areas dealing with legal requirements and restrictions, data flow from collection to disposal, and the parties that may be involved. You will need to identify additional topics and further develop the questions surrounding those topics. But from those responses, you want to be able to assign some type of metric, be it a point system for risk, assessing by maturity, or some other type of model. Where you identify hurdles, you then want to consider mitigation measures to proactively address.

[Slide: Design | Development | Mitigation. Train and Communicate: train and communicate on purpose, scope, and limitations on use of personal data collected. Secure: identify security measures aligned to the sensitivity of the information collected. Test: run tests to assure that measures implemented are working properly.]

And some of the common mitigation measures will include effective training and communications, employing security measures, and auditing and testing. Now, some of the things that I've seen that go awry within companies' programs is, for effective training and communications, only focusing on the onboarding process and not continuing to train and keep things top of mind throughout the course of their employment. So those are little things that, even if you do have current policies and practices in place, you can tweak or challenge or get creative with to make sure that training and communication, security, auditing, and testing are being employed in a timely manner and thoroughly.

Then when we get to the data collection stage, this is about defining the relationship between you and the data subject through proper notice. But you can't provide that notice without really understanding the data lifecycle for the particular project at hand. So what are the parts that make up the data lifecycle, and what are you trying to understand? First is collection. How are you collecting that data?

Are you collecting it voluntarily through the data subjects themselves? Are you getting that information from third parties? Are you going through the web and scraping that data? And what type of data is being collected, and how much? Is it sensitive personal data or just routine personal data?

Is it business contact information? All of those have impacts from what you're going to have to disclose. What are you going to do with that data? How are you going to use it? It may seem like a very simple question, but oftentimes I find that due to silos within organizations, that use question is not always easy to decipher.

And who will it be shared with? Are you going to involve third parties? And if you do, are there agreements in place? And do those agreements take into consideration privacy? What can that third party do or not do, or [inaudible] prohibited from doing with information that is being collected?

How long do you need to hold on to that information that you're collecting, and why? And when it's finally time to dispose of it, how are you going to do that? Is it going to be through physical destruction? Are you going to sanitize that information, meaning you're going to anonymize it or pseudonymize it as it's permitted under the GDPR? What steps are you going to take?

Once you can answer these questions, then you can move on to the step of drafting an appropriate notice that will inform the data subjects, which will lead to informed consent.

Now that we talked about what you need to know in order to provide that notice, how and when do you have to provide that notice? Well, first, make sure that you use plain language, or the native language of wherever you're going to be employing that practice, and avoid being overly lengthy, legalese, or complicated. Be very, very clear, and provide that notice prior to or at the time of collection. Then collect that consent, and make sure that you have some way of maintaining a log of gathering that consent.

[Slide: Archive | Disposal: Store the data collected while it's useful or legally required, then dispose of it.]

And as mentioned, when it is time, assure that the personal data and their copies are deleted or otherwise sanitized to no longer fall under the category of personal data.

So now we've discussed why Privacy by Design should be considered by companies. Then we went through the seven principles that make up the framework of Privacy by Design. We went over a high-level overview of the implementation stages. So now let's put it all together. First, what I recommend doing is to create a map to outline the process flow for what you are trying to accomplish.

So let's take into consideration recruiting. In the context of recruiting, you need to start with what is the position being promoted, and what are the criteria that are important to the position. So step one, what is the selection criteria? And then what is your ultimate end goal? There are going to be a lot of applicants, a majority of whom will get rejected, and a few that will get an offer extended.

What is going to happen to the data collected of those that get rejected? But that's the end. And there are many steps in between. And through each of these steps, need to question from a privacy impact perspective. So a sampling analysis may go something like this.

[Slide: Recruitment | PbD Sample Analysis (table of steps, actions and PbD principles). Application Form: determine application requirements, assess need for personal data (Privacy as a Default). Applications (Online): notice and consent (if necessary), use plain language, do not pre-check, encrypt transmission, third parties/vendor agreements (Visibility and Transparency; Respect User Privacy; End-to-End Security; Full Functionality). Applications (Handed / Referred): notice of use and link to privacy policy via email acknowledging receipt (Visibility and Transparency; Respect User Privacy). Routing of Applications: access limitation (data sharing), implement security measures (storage) (End-to-End Security; Full Functionality). Rejected Applications: rejection letter (disclosures), disposal of application materials (Privacy as a Default; Respect User Privacy).]

Under the application form, are you going to do it paper, or are you going to do it physical? And what type of information are you going to collect through that application? And make sure that you really understand the need for it. Are you just asking for it because, or do you really need it? An example of this is a common practice that we run into within the HR field through the recruitment, which is to ask for references.

In your process, are you really going to be using those references at the beginning of the stage? Or are you only going to be contacting references if you are closer to making an offer, or a decision on making an offer, to a candidate? If you only need the references from those that are making it past further in the stages of the recruitment process, then collecting at the beginning doesn't make a lot of sense from a privacy perspective. So again, challenge some of those practices that you've had in place. And then for applications, do you need to gather consent?

Is it going to be opt-in or opt-out? What security is employed, and are there third parties who are going to assist you? And you can see how these map to the Privacy by Design principles. Again, this is just a small sampling of analysis that needs to take place under the Privacy by Design framework, but once it becomes part of the culture, it will get easier. And people will start to understand what they need to think about and the impact of those answers.

Implementing a privacy program is not easy, but you just need to take it one step at a time. Considering a Privacy by Design framework is a great place to start to understand your obligations and designing workflows so that privacy is considered proactively rather than retroactively after an event. My contact information is on this slide, and if you have any follow-up questions, do not hesitate to reach out. Thank you for joining us today.

Questions answered in this webinar

What are the main stages of implementing Privacy by Design?

At a high level there are four stages: purpose identification, design and development, data collection and use, and archive and disposal. Each stage calls for systematic questions about the privacy impact of what you are building. Answered by Soo Kang.

Should a recruiting process collect references from every applicant at the start?

Only if you will actually use them then. If references are contacted only for candidates close to an offer, collecting them from everyone at the beginning doesn't make sense from a privacy perspective, so challenge long-standing practices like this. Answered by Soo Kang.