Facebook defends ad model, bigs up its power for good, still squirms around GDPR

Stuart Lauchlan Profile picture for user slauchlan April 25, 2018
Facebook's Zuckerberg and Sandberg couldn't ignore that data privacy elephant in the room, but didn't provide much fresh clarity either.

Sandberg and Zuckerberg

Earlier in the week Alphabet boss Sundar Pichai managed to get away with not talking to Wall Street analysts about data privacy issues in the wake of the Cambridge Analytica revelations igniting a cross-industry debate on the subject.

There was no way that Facebook’s senior management was going to get that same benefit and that’s how it turned out yesterday when the firm turned in its Q1 numbers. That said, those numbers were strong enough for Wall Street’s finest not to rock the boat too much…

But Facebook CEO Mark Zuckerberg was at least asked what he’d learned from his recent Congressional (very light) grilling. His response was a well-rehearsed reiteration of the party line that’s become very familiar in recent weeks:

I think that that was an important moment to be able to go and hear what people were wondering about and just to have a public hearing of answering all of the questions around Cambridge Analytica and what we knew and all the steps that we're taking on data privacy and developers to make sure that this doesn't happen again and to lay out all the different things that we're doing.

I mean, the hearings didn't just touch on that. They also touched on a number of the other issues that we face, including foreign interference in elections and that's something that we're incredibly focused on. 2018 is going to be an incredibly important year on this. There are big elections, not just the U.S. midterms, but the major elections upcoming in Mexico, in Brazil, in India, and Pakistan, and a number of other countries around the world.

So this is important and it was an important moment for the company to hear the feedback and to show what we're doing. And now I think the important thing is that we execute on all the things that we need to do to make sure that we keep people safe.

Put like that, he sounds almost grateful for being the center of attention of one of the biggest scandals to hit the tech industry! But it was front foot forward all the way in pitching the ‘mea culpa but…’ messaging:

For most of our existence, we've focused on all the good that connecting people can bring. But it's clear now we didn't do enough to prevent these tools from being used for harm as well, whether that's foreign interference in elections, fake news, hate speech, or app developers and data privacy. So now, we're going through every part of our relationship with people and making sure we're taking a broad enough view of our responsibility, not just to build tools, but to make sure those tools are used for good. This means continuing to invest heavily in safety, security and privacy.

We have a responsibility to keep our community safe and secure, and we're going to invest heavily to do that. At the same time, we also have a responsibility to keep moving forward and keep building tools that bring people together in meaningful new ways. That's what makes Facebook so important to so many people and that's our responsibility too.

There have been major pluses delivered by Facebook, he insisted, citing Artificial Intelligence tech as a powerful enabler for good:

AI is the most important technological trend right now and I'm optimistic that it can help us amplify the good that's happening on our services as well as proactively remove harmful content. For example, one thing that I'm proud of is our AI tools that help us take down ISIS and Al Qaeda related terror content, with 99% of that content being removed before any person flags it to us.

We've also built AI tools that have flagged when people are posting thoughts about suicide. And these tools have helped us reach out to first responders to get over 1,000 people the help they need quickly. On the positive side, AI will help us understand the context of what people are sharing so we can help encourage more connection and conversations between people as well.

No ad impact...yet

The Cambridge Analytica mess hasn’t had any impact to date on Facebook’s ad revenues, although clearly this may change over the coming quarters unless the company is able to put the genie back in the bottle. But for now, said Chief Operating Officer Sheryl Sandberg, it’s pretty much business as usual:

In the immediate days of the concern, we heard from a handful of advertisers who paused spend, one of whom has already come back, and we haven't seen a meaningful trend or anything much since then. Advertisers ask the same questions as people are, that they want to make sure their and their customers' data is protected, and I think we are able to answer those questions in a compelling way.

There’s nothing to be ashamed of in Facebook’s data-dependent ad model, she insisted:

We're proud of the ad model we've built. It ensures that people see more useful ads, allows millions of businesses to grow and enables us to provide a global service that's free for all to use. The fastest way to bridge the digital divide in the United States or around the world is by offering services free to any consumer regardless of their circumstance. Advertising-supported businesses like Facebook equalize access and improve opportunity.

At the same time, we know that people want control over how their information is used, and we want them to feel confident that the ads they're seeing are authentic. That's why we're building industry-leading transparency tools. This includes a way to see ads an advertiser is running, even if they aren't targeted to you. This new feature is live in Canada and will roll out in Ireland and the U.S. soon.

GDPR shifting

As for Facebook’s shifting stance on GDPR (General Data Protection Regulation), anyone hoping for clarity should consider themselves still hoping. After Zuckerberg said that non-European Union users shouldn’t expect the same level of protection, then rowed that back, then added in a few more ‘ifs, buts and maybes’, Sandberg had a stab at her interpretation - and it’s all about opportunity here, it seems:

In the coming months, GDPR will give us another opportunity to make sure people fully understand how their information is used by our services. It's an EU regulation, but as Mark said a few weeks ago, we're going to extend these controls to everyone who uses Facebook regardless of where in the world they live.

OK, so that seems pretty unambiguous. But hold hard, what’s this? Sandberg added:

It's not going to be exactly the same format. It's going to be localized instead for different parts of the world. And so we think some of the differences will come from that.


Obviously the issue for Facebook is that GDPR’s protections potentially have a negative impact on its ad model, an issue which Sandberg was ready to concede:

When you think about the way people have the choice to restrict data use, I think it would affect the product. There are lots of ways we use data to make the product better. It really depends what that would be. I don't think we have full visibility into what those changes would be over the long time…The way we think about it is that the amount of uncertainty there is for us, and all the other companies in the digital advertising industry, is reasonably higher than it's been right now because we're in the process of rolling out GDPR.

My take

So, on GDPR that’d be ‘we haven’t a clue how bad this might be for us’ then? It’s worth noting that Chief Finance Officer David Wehner is rather more frank here:

While we don't expect these changes will significantly impact advertising revenue, there's certainly potential for some impact. Any change of the ability for us and our advertisers to use data can impact our optimization potential at the margin, which could impact our ability to drive price improvements in the long run. So we'll just have to watch how that plays out over time.

And he acknowledged that the Facebook pitch to advertisers could be hurt by GDPR:

I don't know that we really see a doomsday scenario here. I think what we think is that depending on how people react to the controls and the ad settings, there could be some limitations to data-usage. We believe that those will be relatively minor. But depending on how broadly the controls are adopted and set, there is a potential to impact targeting for our advertisers.

Obviously, if they are less able to target effectively, they'll get a lower ROI on their advertising campaigns. They'll then bid differently into the auction. That ultimately will flow through into how we can realize price on the impressions that we're selling.

Quite - so it's potentially a really big issue for Facebook, no matter how much obfuscation takes place.

Overall, another day, more questions for Facebook. Another day, no more definitive answers on offer. For how much longer can this go on? For as long as the ad numbers hold up, I imagine.

A grey colored placeholder image