Targeting in the New World of Privacy

As we use the internet more in our daily lives, keeping our online privacy has become a big concern. People want to be safe online, so there are now more rules about how companies can use our information. This includes making the internet more secure, getting rid of some types of cookies, and being more careful about how we handle digital information. These rules are good for people, but they make things a bit tricky for businesses, especially in marketing. To succeed in this new online world that puts privacy first, marketers need to come up with smart and new ways of reaching people.

The next guest post series aims to give marketers tips on how to best behave in this new world of privacy. We asked the leading companies in digital advertising these important questions.
 

In what ways do you feel that privacy-related changes impacted (your) targeting strategy? What are your recommendations for the right Targeting Strategy in the new world of privacy?

I believe we can identify different phases in online advertising, when it comes to targeting.

In its infancy (and here I’m talking late 90s), targeting in digital media was not that different from offline media. People would reach out to the publishers, ask about their audience and, if there was a perceived fit, would buy some media space.

Cookies changed the whole game. As tracking across websites became a thing, real niche audiences also came to life. Then we moved to what we ideally consider “digital marketing targeting,” where one identifies super specific niches. Think of it as “Male, 25-35, expat, who runs, likes football and plays crossword puzzles”.

But there is a step beyond that, which are the “data-driven audiences.” Until very recently, the best practice on platforms such as META was to just upload your audiences and let the platform itself find the commonalities between the people who are actually taking the actions you want on your website. META allegedly used 98 unique data points to define advertising audiences (I personally believe it was way more). Still, the fact is that there is no way one could come up with and implement such granular data without machine learning.

This is all to say we were already living in a “post-targeting” digital marketing world.

These recent privacy changes came in to disrupt all of that. The number of data points that platforms have definitely will decrease, so there are fewer data points for algorithmic targeting, which makes lookalike audiences less effective. But it’s not like you can just go back to do things manually. Because your previous audience of “male, 25-35, expat, who runs, likes football and plays crossword puzzles” is now not identifiable either (without consent, there is no way to build this segment).

Ok, so how do I believe this changes the game?

The first one is how central creatives become. Let’s be honest, when algorithms were doing all the targeting, you could survive with subpar creatives. The machine was so good at delivering ads to the right people at the right time that creatives became a secondary issue. But now, a creative is the targeting. When you launch a new ad or campaign, platforms will show it to a broad number of people, identify who engages best, and then try to refine this “top engagement” segment. For this reason, your creative needs to be engaging with the audience you want to target. It’s probably the most crucial aspect of digital advertising.
Now, the second thing is about data. You need to establish a good relationship with your audience, so they will willingly give you data and their consent to share it with third-party vendors. However, this is a tall order, after all, the whole situation is that people don’t want to share data in the first place. For digital marketers, this raises the problem of creating a “reason to share.” You need to create a trustworthy brand and good enough content or offers that people are willing to exchange it for their data.

The last thing is about experimentation and measurement. Before, you could do proper AB tests with audience segments and see how they performed. Now you can try different approaches, but because tracking is less exact, it will need more sophisticated measurement methods. That is, if before you could say, “Audience A converts better than Audience B,” now the best you can do is “When we do X thinking of audience A, we see a conversion uplift.”

How can marketers make sure they follow the rules about privacy while still making ads that feel personal and work well?

The only way to make great ads is to make a lot of them. I believe that the way forward is constant testing and optimizing towards business outcomes. As I mentioned before, you might not be able to reach a super-niched target audience anymore, but your goal should be to create great ads with that audience in mind, so it resonates.

The key is finding the right hook to capture your audience’s attention.

My suggestion here is to broaden your scope when researching. Of course, dig into your own data to develop insights, but don’t constrain yourself. I believe you should be reading the comment section of your (and your competitors!) ads, scrolling through review websites, and finding forums where people talk about the problem you are trying to solve. This is where you can find good creative inspiration, and has nothing to do with privacy regulations.

As the privacy rules change, what factors should marketers think about when picking the tools to advertise via, so they keep our information safe and still reach the right customers?

Marketers don’t pick where to advertise, their audiences do. So the first step to finding the right places to advertise is to understand where your target market is.

This rule already weeds out a lot of conflicts regarding information practices. For example, if your audience is extremely privacy-focused, they are probably not on META anyway, so it makes no sense for you to advertise there.

Now, when you expand your marketing tool kit, the first step is a data assessment, and checking if your company is comfortable with how the data will be handled.

Aside from that, I think it’s an issue of trust and transparency. First and foremost, users and customers need to be able to trust that your company has their best interest in mind.

How can marketers use the right technology to make ads that reach the right people without being too nosy about their personal data?

One of the main changes that marketers need to be comfortable with is not knowing. As I mentioned before, the new privacy regulations are shattering the illusion of full visibility on user journeys and attribution. What we can do, however, is to take a data-guided approach to ad creation. I mean, run multiple ads, and see which ones are really achieving their objectives and moving your business’ needle.

There are two ways of doing it. One is having a deterministic approach, and just comparing creatives based on their impressions, clicks, and conversion rates. This can tell you a lot about what is working from a practical standpoint. The other approach, definitely more complex, is trying to run some correlation analysis and marketing modeling. You can start tackling questions like which message or creative brings more incremental results or which one works with people who don’t know your brand.

Regardless of the approach you take, the underlying need you will have is clean, workable data. If you are talking about tools, I recommend Funnel, which can give you creative-level data from all platforms and help you answer these questions.

Why is it important for marketers to be open and honest about how they use data, and how can they talk to users in a way that makes users trust them more?

I believe that at some point, data privacy and transparency will just be expected from any legitimate business. In the same way, people assume that brands have the highest possible standards to deal with their credit card information, for example, they will just expect that any legit business handles data with privacy in mind. However, we are not there yet, so if you want to be cynical about it, you can argue that being upfront about data and how it’s used is simply a competitive advantage.

But it’s also an aspect of brand building. I can’t think of a single marketer who would say NO to associating their brand with trustworthiness, and being clear about data and privacy is a way to establish that.

Now, when it comes to communicating with people, the key here is transparency and agency. You should be clear about how and why the data will be used and allow people to change their minds over time, or customize what data will be shared.

Did you find cases where new marketing methods of the post-privacy world actually improved your performance or the way you interact with your audience?

For a long time, marketers had the illusion of accuracy. We had this idea that we could track every single click and have the complete picture of the consumer journey, attributing conversions to each ad people saw along the way. It was a nice idea, but it was smoke and mirrors.

Things outside the marketer's control always influenced people, they always had messy, non-linear journeys, and all we got were just models, not truths.

The main benefit of this shift towards privacy is that we have shattered this illusion, and this is forcing us away from a deterministic worldview. The marketer's job is becoming more sophisticated, where we are working with correlations and probabilities. We need to be less of media buyers and more of strategists.