E14: The Three Pillars of GDPR (🎧PODCAST)

The three pillars of GDPR are transparency, security, and consistency. Regulators have stated they see transparency as the central point, but the other two are equally important. When a natural person asks your company for a Data Subject Access Requests (DSAR), how will you react?

The Ward brothers explain what companies need to expect within the three pillars and what is likely to be the biggest burdens as the GDPR rolls out this week.

TRANSCRIPT

Jay: “Are You Data Smart?” Weekly podcast on data security, information management and all things related to the data you have, how to protect it and maximize its value. I’m Jay Ward.

Christian: And I’m Christian Ward. Today we’re going to tackle the three pillars to GDPR. Three keys, three H1 tags, I don’t really care what you call it. The three most important things that I think honestly, Jay, a lot of people are writing about and talking about. We’ve pegdemised [SP] transparency security and consistency and we’re gonna walk through each of these. We’re also gonna make sure that we’re blogging about them at wardpllc.com each of the next several days just to sort of send off the voyage of GDPR on a well-christened voyage.

So as we approach that, Jay, let’s jump in right away in terms of you know, sort of thinking about transparency because there is a lot of miscommunication right now regarding privacy policies, what’s out there, what are people really updated. I think a lot of companies in the United States don’t really even know if they need to make updates. I can not tell you how many new privacy policy notifications I’ve gotten in the last 16 hours alone. It is blowing my mind. I wanna like clip them all and do sort of the, you know, what’s that meme where it’s like you know, that guy walking with the girl and then he looks back at the other girl who looks exactly like that girl and it’s like your old privacy policy and your new privacy policy because, in some ways, they are identical anyway. I just feel…

Jay: I can think of the exhibit, you know, I heard you like privacy policies so I put a privacy policy in your privacy policy.

Christian: Yes. It’s a little disturbing. So I will put that into one of our blog posts shortly because that is rather funny but let’s talk about you know, starting with transparency. Like give me some of the thoughts you have around, not just, you know, privacy policies, which need a massive overhaul, but why and how does GDPR really drive us there.

Jay: Well I’ll tell you that I’ve looked at, I guess, you know, maybe one or two privacy policies and in the course of doing so, I’ve noticed a trend which is there was like a period in 2008 when privacy policies were revised and just everybody copied them. They all looked the same. You could run a red line and the difference would basically be, you know, the name of the company and the address and pretty much everything else would be the same even when there were stuff that made no sense like you’d have a retail store talking about, you know, getting disclaimers for using helicopters and it was ridiculous. It was that sort of shortcut thinking copy paste stuff that first-year lawyers do that drives everybody crazy.

Christian: So it’s like the MS Word paperclip popped up at the bottom of the screen..

Jay: It looks like you’re writing a privacy policy.

Christian: It looks like your writing privacy [crosstalk 00:03:03]

Jay: Clippie. Yeah. Clippies had a resurgence as a meme themselves actually.

Christian: I know. I’m happy for them. He was the first AI.

Jay: You know he needs it. You know he really deserves some credit but get this…now, we’ve got the second tranche of privacy policies coming out and I actually see a divergence. There seem to be two kinds. There is the one that looks very different that says we use your personal information. We use it so that we can improve our service, so that we can communicate with you, so that we can make our products better and we share it with these people, here is why we share it with these people, here’s what we’re gonna do with it, you don’t have to allow us to do that if you want to object you can do X, Y or Z, contact us here, call us here. And it’s very straightforward and layered and that’s what a GDPR compliant privacy policy is about.

The second type that I’ve seen looks exactly like the old ones. It’s about 20 pages long and nine-point font. There is “where” fors and “here to” fors and you know, it’s like Ben Affleck, you know, your suspect. It’s ridiculous. And there’s no difference. And that’s gonna come across the desk of data protection authority in Bossols or Dublin or London and they’re gonna say, “Oh, no, I don’t think so. I didn’t like ‘Good Will Hunting.'” So you can’t do the same thing that we were doing before the GDPR and expect that it’s going to work. So the reason for that is because the GDPR’s approach to all of this is enforcing transparency and we’ve blogged about this and talked about this before. The Data Protection Authorities, the regulators themselves have said that transparency is the most important principle that they are going to be looking for when they are making their enforcement priorities. They are gonna look to see if you’re telling the truth about what you’re doing. If you’re being straightforward and clear about what you’re doing. Because there’s no way for a data subject to exercise her rights if she doesn’t know what data is being collected and why. So…

Christian: Yeah but the…Let me stop you. The reality is this, so I just made a joke earlier of you know, I’ve gotten a thousand of these notices, I guarantee everyone listening is getting them as well so you’re getting all these notices. I’m not reading them, I mean, I get it, you’re a lawyer, you have to at some point but I’m not. And so the Regulatory Authority is saying we now have to disclose what you’re gathering, why you’re gathering it, how you’re going to use it. How do you see that playing out? Is that that, you know, 50 people get together and decide to bring a class action suit because they think that you’re using their health fit bit, you know, Garmin type of tracking data in the way you shouldn’t. How do you see this actually coming to fruition where these new regulations do in fact make a change.

Jay: I see it as being sort of a dual approach. There is the top-down approach which is the enforcers are going to look to see if companies are obeying these requirements. They’re going to be making sure that there’s nothing that appears in public. There’s nothing that you’ve done. There’s been no breach that triggers a finding that you were engaged in, you know, non-transparent activities. But even more than that, it’s really about the threat of that happening. The threat of 4% of Annual Global Turnover should be enough, or in many cases, 2% of Annual Global Turnover. That should be enough to keep many companies in line. It’s like in “Star Wars,” fear is going to keep many businesses in line and it will do so because the risks here are so serious. So that’s the top-down.

The bottom-up aspect of it is, you don’t know when someone’s gonna file a data subject access request with your company. You don’t know when someone’s gonna file a complaint with the Data Protection Authority and the DPAs have to act on those. They have to act on every complaint they get. Companies have to act, with some limited exceptions, on every DSR that they get. And if it turns out that the DSR reveals that you were processing data way beyond the scope of what you disclosed or if the complaint reveals that you were processing data, sensitive data without getting explicit consent, you’re in for a world of trouble. So there’s both the, you know, the implicit threat of enforcement and the governmental approach and then there’s what in this country we often call as private attorneys general, like, individuals are empowered under the GDPR to be a check on what companies are doing with by demanding their access to their data. And if you don’t think there’s at least one person who’s a customer of every company in the world to be like, “You know what, I’m gonna file a data subject access request. I’m feeling up to it.” There’s one guy who’s always gonna do that.

Christian: Yeah. Look, I think that’s, I think that’s spot on. You know we talked a little bit last time about you know someone creating an app that you just enter your email address. It’s really like the gentleman I’m not sure if he’s from Microsoft where he built the platform to check if your password has ever been breached by looking at your username because he can check online to see if you’re username was there. It’s a very smart solution but it’s a similar concept. It’s like a do-not-call checklist and or unroll me. We tweeted about unroll me a while back where it’s sort of unrolling from newsletters. Stuff you don’t even know you’re connected to. But this goes so much further, you know, asking for those, you know, the reports on what data someone has. I do think we’re gonna see mass adoption of that. I think we’re gonna have entire populations of people filing these through a joint website and, you know, not trying to give anyone ideas out there but this is definitely gonna happen if it’s not already in the works, and when it does, you know, I think to your point, having a policy already laid out that allows for people to understand why, how, what you’re using it, that is the point of the privacy policy. So we’ve moved away from Clippie the MS Word auto privacy policy generation into something that really does need to state, not necessarily because people are gonna read it upfront but surely because everyone’s gonna read it after the fact.

Jay: Yeah that’s right. And you know…the guy you’re talking about Troy Hunt from “Have I Been Pawned?”

Christian: Yeah yeah. Exactly. It’s Troy Hunt. Absolutely.

Jay: And he is in Microsoft. His efforts, I think, really do go to the second pillar which is security. And that’s embodied for me, I mean pretty explicitly in Article 32 the GDPR which is about security processing. And this is the one that is, I’ve talked about this before, it’s the easiest to define and the most difficult to do. Because what the requirements of Article 32 are are you have to take into account cost, the difficulty of implementation, state of the art, the likelihood of risk. All of these various factors into coming up with what is the appropriate amount of security for the type of processing you’re doing, for the type of data you have. And that’s it. That’s what it says. You have to do that and that’s the end. Well, that’s a straightforward requirement that most people could understand. It’s doing it that’s going to be difficult because how do you know if your security measures were adequate enough unless you’ve been breached, you don’t.

You know, there was a Calvin and Hobbes when Calvin asked his dad, “How do they know that the weight limit on that bridge is 25 tonnes?” And his dad says, “Well they drive progressively heavier trucks over it until the bridge collapses. And then they know that’s how much the bridge can stand.” Well, data security can, in some ways, be like that, you know, you only know how good your data security protocols are when you’ve either successfully repelled something you know by the skin of your teeth or you had a breach and that’s why [crosstalk 00:10:42]…

Christian: Mister OWL, how many licks does it take to get the center of a Tootsie Roll Tootsie Pop Same style, process.

Jay: How many group force attacks does it take to crack your password?

Christian: Yeah, but you know it’s so…And that is really important, it’s…Look I think I take the perspective generally from a data background that most breaches are inevitable. It’s sort of like when we used to teach Pre-Cana classes to couples that were all excited to get married and I would kick off my session with, “I’m so excited to see all these great couples here today. Please look to the couple to your left and look to the couple to your right. Say hello.” And they would do it. And then I would say, “If in five years either of those couples aren’t divorced then you are.” That’s the reality. That’s why…

Jay: That’s why you were not asked back to Pre-Cana by the way.

Christian: Yeah. Well, it’s like the epitome of getting out of jury duty. But any…the point being the statistics don’t lie. Most companies, there’s something like eighty-something percent of companies are expecting that this is going to happen to them or it already has happened. GDPR is obviously getting a lot of people to rush and tell the world about it before the next few days May 25th but my state mode be is that there isn’t enough security in the world. You cannot stop your own employees which I believe is probably still the most dangerous thing as your employee is walking in…

Jay: By far.

Christian: Yeah. And walking out. There’s nothing that can stop them.

Jay: Or negligence.

Christian: Yeah. And look I think that’s the point right? So any sort of regulation that says you have to have adequate or as adequate or as necessary it’s gonna look like it paled when the actual breach occurs and they’re gonna go, “How come you didn’t see that coming.” It’s just very hard to fight. I think the important thing in security here is similar to everything we talk about with regulatory authorities is documenting your process, documenting your system, your understanding of the risks and laying it out there. I know it sounds counter-intuitive, in some ways to say, “Hey, if we document all these and then it doesn’t go exactly according to plan.” But that’s the world we’re living in, you do need to have at the very beginning of each of these discussions the security by design, or the privacy by design by taking to account how all these different groups interact with each other. It’s not just the CTO’s or the CIO’s decision and what to do with the data is not just the CDO’s decision. It really has to bring all those teams together.

Jay: Right, I mean, you know, if you’re talking to about a breach, you need to be able to identify and report an incident as quickly as humanly possible and that means down to your CSRs or the people who are working in your retail store. I mean you need someone to educate your entire team about what it means to be GDPR compliant and that’s the security component. But it also dove tails with the consistency and that’s, I mean, that is the ultimate question here. That is the long-term…

Christian: You mean the third pillar.

Jay: Yeah, the pillar is consistency. How are you going to do this over the course of the next 20 years? Because I can tell you it’s unlikely that there’s gonna be as big a change in data security in the next 20 years as there has been with the GDPR. I mean, the last thing we had was the ePrivacy directive in the mid-’90s and it basically was not very effective. And, you know, now we’re talking about…

Christian: You know those little cookie banners, they’re amazing. I mean, think of all the lives saved. You know, yes, I don’t think it did enough. I am actually decided to see what ePrivacy, you know, the regulations now are going to do. I know it’s too late but you know ultimately that is another good front line. I think, you know getting that, not just the banner but a much better understanding of what do cookies do, what are they tracking, what do you allow them to track. This is gonna be that age-old discussion of it’s my data to some degree of about myself, my first-person data, what I share with you. What value do you place on that? Instead of me giving it to you for free, unwillingly, unknowingly, now hopefully I’m going to get some information as to what you’re doing with it.

Jay: That’s right and ultimately the critical component of consistency is that it applies at every level. You have to be consistent in implementing privacy by design. You have to be consistent as Article 30 says, in documenting all of your processing activities. You know, and that doesn’t necessarily mean I need to know, need to disclose to the public and to the regulators, you know, on this day my team was looking at the metrics on these four call centers about how to optimize the second part of the channeling function. Like that’s…it doesn’t have to be that granular because that doesn’t mean anything.

Christian: That is a good question though. So, you know, I work with a lot of different analyst teams over the years and so one thing is there’s sort of record keeping, there are outlines. When you’re taking records in this manner, the question I always come to is, if we have 15 data sets, let’s say we’re just trying to understand the responsiveness inside of a chat window for a customer service representative to getting to an answer or a proper disposition of a question. And there’s a lot of ways to analyze that. You can do the time, you can do the semantics around the actual questions, you can try and type back to a knowledge base. We come at those problems probably 50 different ways in the four hours we look at the data across five analysts. What do we need to disclose there? Because there’s definitely PII in there. You would be amazed what people would tell you in a chat window and there’s a lot of other things going on with that information. What do you think is fair? What protects you in being consistent but isn’t so onerous to the fact that, yes, it is how I’m processing it and I’m using it for certain purpose but is it enough for me to just say we’re just trying to get to a more efficient better customer service experience?

Jay: I love that you asked the lawyer a question about threading the needle like what’s…what works in this particular service [inaudible 00:16:30]. What did you think was going to happen here?

Christian: Sorry.

Jay: The answer is, of course, it’s dependent on the circumstances in the facts and that’s I think what the regulators are gonna say, too. You disclose what you’re going to do to the data subjects. You explain what you’re going to do to them and that’s the transparency. You protect the information and you ensure that it’s not accessed by those who shouldn’t access it, that’s the security. And you only process the data in accord with what’s been disclosed and what’s appropriate under the legitimate basis and the lawful basis you have for processing. That’s consistency. But when it comes to disclosing what exactly it is you’re doing, you need to think about all three, right? Because if what you want to do is make sure that the activities that you’re engaged in are protected, you need to think about what is an honest assessment of what it is you’re really doing. What am I trying to do with the processing of this data? And if the answer is 10 different types of ways to optimize the call or to optimize the user experience in a chat window, if you can say we use your data to optimize your experience and chat windows, I think that’s…that to me in many ways seems okay because it’s an honest appraisal. It’s an honest statement of what you’re doing. They know the data that you’ve used to get there. And as long as you don’t start using the data now to determine, you know, geographic concentrations of buyers for a particular kind of direct marketing, which is outside the scope of what you’ve disclosed, I think you’re okay.

Christian: Yeah, but I wanna do that next week. So I was just talking about today. So, yes. Yeah.

Jay: Welcome to the life of a Data Protection Officer.

Christian: That’s right, that’s right. Now I could look at…I think if the reality of it is that we are talking about as best you can, you know, fitting the curve and making sure that we’ve got enough information to protect customers, give a good explanation of what we’re doing with the data and reporting it back, then I think that obviously makes sense.

We’re gonna wrap it up there. That’s us talking through what we called three pillars of GDPR: transparency, security, and consistency. We look forward to next week. We’re gonna dive further into some of the concepts I wanna read, it’s at Article 22 which is the automatic decision-making process as it relates to each of the three pillars, and look for some more articles by Jay just before May 25th at the wardpllc.com blog where he’s gonna discuss each of the pillars in more depth. Thank you again for listening to “Are You Data Smart?”

Jay: Thanks again.

Leave a Reply