The Future of Privacy in Digital Advertising

Joe Mineo
October 8, 2021

Advertising and analytics have changed drastically in the past few years, especially in 2020. This series, “The State of Advertising and Analytics,” explores the evolution of media, advertising and privacy, the rapid growth of data analytics in marketing, and the future of marketing and analytics, providing useful tips and value along the way. Read the first chapter here


The whole point of advertising is to reach people interested in buying your product effectively, and if that channel is valuable, putting more money in that direction. But where do we draw the line between effective targeting and encroaching on consumer privacy? This chapter deals with the two areas in which digital marketing has both moved the entire field of advertising forward, but also moved it back a few steps: targeting and attribution.

Hypertargeting: the new norm

First, let’s talk about targeting. Digital ads are unique with how they can reach people, in that most platforms use a healthy amount of owned or purchased data to generate an audience. Instead of relying on unreliable demographic surveys, walled gardens like Facebook and Google — places that collect data on their users and don’t share that info with partners — realized they could use their trillions of data points to make more effective advertising, and it caught on like wildfire.

Businesses don’t have to guess that they’re reaching women 45-64, those users provided that data when they signed up and agreed to the platform’s terms of use. Facebook learned that certain people liked and commented a lot on soccer content, so they use that info to serve those people ads related to soccer. Harmless enough, right? Well, the internet can’t have nice things for too long before bad actors find ways to spoil it.

Companies of all sorts realized that Facebook and Google were making heaps of money, so why not cash in? Experian, Oracle, Transunion — the credit and HR companies that we rely on for buying houses, cars, or providing us with health insurance — started packaging your information and reselling it, which helped Facebook, Google and a lot of other ad platforms get scary accurate on their targeting. Even without the Cambridge Analytica scandal, targeting would have eventually reached a critical point where they’d put their foot down to demand change.

Attribution retribution

Next, let’s discuss attribution, which is a fancy way of saying how businesses assign credit to an ad or channel for completing a task. For example, if someone buys my lawnmower specifically because they saw a Facebook ad about my lawnmower, I would assign credit for that sale — or attribute the sale — to Facebook. TV, radio and print are very difficult to attribute, but with tracking cookies, companies like Facebook and Google can follow your clicks and actions to show their value.

Again, these companies struck gold, because they could prove with extremely high certainty that they were the reason businesses were making money. If a CEO has to decide between $1,000 for a radio ad with an estimated ten sales, compared to $1,000 for a Facebook ad with 100 officially tracked conversions, we all know what horse that CEO is going to bet on.

However, just like with targeting, people got greedy and started using these behavioral data points — likelihood to purchase, interest in one product over another, high chance of clicking the “contact us” button — to create a monster beyond our wildest dreams.

Sure, if I’m interested in soccer that’s one thing, but showing me ads based on digital or even external behaviors, knowing that if I see it on a Tuesday when it’s partly cloudy with 60 percent humidity because they know that’s when I purchase most, has taken it to new levels. Companies sprung up overnight to collect, connect, package and resell this information, creating a huge marketplace to target people directly.

Something to consider is the transfer of information from hardware and internet service provider (ISP) companies (think Apple for iPhones, Samsung for semiconductors that go into smartphones, or Comcast/Verizon for internet) and advertising platforms like Facebook. Each phone, laptop or tablet has a unique device ID, and each device links to an IP address mapped to an internet connection. This information can be linked back to phone numbers, which are tied to home addresses, voting records and, to bring it full circle, credit reports.

When behavioral attribution data gets married to the targeting data, privacy all but disappears. Introduce machine learning to fill the gaps, and you could bet that some of the best platforms out there could pinpoint anyone in the world with solid accuracy. 

The internet begins to self-regulate

Now that we have a base understanding of targeting and attribution, let’s move on to the evolution and regulations introduced to combat this privacy issue, along with how advertising will likely respond.

If you’re not familiar with Europe’s GDPR, Google’s move to kill third-party cookies, or Apple’s iOS14 changes, here’s a brief rundown in non-technical terms. General Data Protection Regulation, or GDPR, is Europe’s way of forcing companies to disclose how much data they track on a user through cookies. A user must consent to tracking in order to be tracked, which kills the efficacy of tracking cookies at scale if a majority of users opt out, and reduces the ability to attribute ads to conversions. In 2020, Google decided to eliminate tracking cookies by 2022, but has since pushed that move back to 2023.

This would heavily impact ad targeting for pretty much every platform in the business, since they all rely on cookies in one way or another, and they would have to create new solutions to attribute and target effectively. Apple, seizing on the wave of privacy progress, not only forced advertisers to tell people when they were being tracked with iOS14, but released “Hide My Email,” “iCloud Private Relay,” IP address masking features and options to kill cookies altogether on iOS15. These three major changes in the past five years have forced the advertising industry to think hard about how they target, who they target and how they attribute success.

GDPR, outside of levying fines against Facebook and Google, has largely been a blip on the radar, mainly because of loose oversight and simple workarounds used by major companies. How many times have you visited a site that says, “If you continue to use this website, you agree to be tracked by our marketing cookies”? Some websites make it difficult to opt out, and the frustration leads users to simply opt in so they can view or download content. These workarounds have allowed advertisers to keep a status quo, and have not largely helped the privacy issue at hand. To say the least, GDPR hasn’t significantly changed or impacted the world of advertising, and other initiatives like CCPA likely won’t either.

Although Apple is a massive manufacturer of digital products, it only makes up roughly 14% of the smartphone market and 31.7% of the tablet market. This is the main argument for Apple’s App Tracking Transparency update, which means that 86% of smartphone users and 68% of tablet users are still able to be tracked through digital advertising in the same ways. Apple’s announcement came under fire from Facebook because they claimed it would impact small businesses unfairly, but the sad truth is that Facebook simply shifted their spending models to serve more to non-Apple products.

Our ads & analytics team looked at data from May to August 2020 and compared it to May to August 2021, and found that Facebook shifted 30% more dollars over to non-Apple products year over year for conversion-focused campaigns, implying that if they can’t find the data with Apple, they’re going to find it somewhere else. Not only is this problematic, as Apple users are just as valuable as non-Apple users, but it trains the models that run the ads to avoid those users altogether, compounding the problem. Users haven’t left the platform in waves the way everyone seems to think, so from their perspective, it’s scrolling as usual. In short, advertising companies will adjust their delivery systems to make “best guesses” at scale. It might impact success rates in the beginning, but over time it won’t make too big of a splash.

Although we don’t know the impact of Google’s decision just yet, as it will fully roll out in 2023, it’s forced the ad industry to look into other targeting solutions, like contextual ads, IP address targeting (assuming other companies don’t kill that too), or other concepts that anonymize user data. Google commands a much larger market share in their territory, upwards of 65%, so if targeting and tracking for a business is purely cookie-based, the death of cookies could spell the death of that businesses’s perceived marketing effectiveness. Some companies will fold, others will grow, and this will likely be the most impactful move out of the three in the ad industry. Finding a replacement for something so uniquely reliable won’t be easy, and there’s billions of dollars at stake.

The path ahead

This subject is heavy, and if you made it this far, you’re probably wondering where things go from here. The simple, unadulterated truth comes down to data ingestion and analytics.

World soccer star Cristiano Ronaldo made headlines when he cast away a Coca-Cola bottle at a press conference, standing up for his philosophy that you only get out what you put in, something he works with a team on. Marketing, in this light, is pretty much the same. Anyone can buy a protein shake, but it takes a team with dedication to get optimal results, with constant analysis of trends and insights. The attention economy will dictate where eyeballs land, and a businesses’ goal is to capture attention wherever it goes.

Facebook and Google ads are so effective — and will continue to be effective — until people stop discovering relevant content. Scale covers big holes, and it will for a time as these changes slowly affect user experience. Advertising will adapt as it always has, and new methods of measurement, targeting and attribution will emerge. User privacy will hopefully be protected, and ideally, marketing will be equitable, diverse, and continue to be accessible to those who wish to grow their business.

Beyond this? Anyone’s guess. If you’re looking to affordably expand your team, ingest more data and take an objectives-first approach to marketing, stay tuned for the next feature in this series, or reach out directly to chat!