
How Do We Make Attribution Work In A World Of Data Privacy?
How we track, measure and attribute the success of marketing activity puts many marketers in a cold sweat—whether we're talking about top-level marketing mix models dating back to the 1950s or the complex, hyperpersonalized multichannel attribution models of today.
The shift to digital has, of course, helped. With tools that can track customers throughout their journey, the process looks easier on the surface. But the truth is that the landscape has become even more complex. There's a skill shortage in attribution, and data privacy poses even more complex challenges. Data privacy laws already make handling consumer data a minefield, and there is little optimism that it will become easier. A 2025 Supermetrics survey of 200 marketers from around the globe revealed that 57% predict more difficulty in marketing attribution in the future.
Why Attribution Is So Important Today
It's worth emphasizing why attribution is more vital today than ever. Perceived wisdom, guided by the marketing rule of seven, has taught us that customers typically need to interact with a brand at least seven times before they decide to make a purchase. Today, however, the digital advertising landscape means customers interact with your brand much more often. Data compiled earlier this year shows that customers interact with a brand 28.87 times on average before a conversion.
With that many touchpoints, it's impossible to understand your successes and failures without an effective attribution model. Attribution helps us understand how customers interact at each touchpoint and enables us to determine the effectiveness of each marketing method. With analysis, we can see which aid conversion and then decide how to spend money and resources more effectively in the future.
The Challenges Of Attribution
Historically, access to data has been a stumbling block for many marketers. You may be unable to access data because you're on a small budget, which prevents you from accessing the right measurement software, or you may have a team that lacks the knowledge to implement what you have. Or there may be a disconnect between sales and marketing—creating data silos that prevent useful data from being used to make smarter marketing decisions.
Access to data is also changing due to user behavior. Nearly 33% of internet users now use ad blockers, which, along with blocking ads, also block cookies that allow us to collect and analyze user data.
Data quality is holding many marketers back as well. Research by the Chief Marketing Officer (CMO) Council and GfK in 2022 found that 62% of global marketers are only moderately confident—or worse—about their data.
How Data Privacy Has Affected Attribution
Since its implementation in 2018, the General Data Protection Regulation has radically changed how European marketers use customer data. There are currently no federal laws in the U.S. that are as comprehensive as the GDPR. However, laws like the California Consumer Privacy Act have started an inevitable shift toward increased data privacy. Legislation like this makes businesses legally obligated to process data securely and limit how they share or use it with other organizations. That means considering things like data processing agreements, which establish your roles and obligations as well as those of any organizations you share data with. It also means implementing robust data governance—ensuring all of your consumer data is clean, reliable and consistent. While essential for consumers, these are all things that take extra time and resources for businesses to implement.
There are also other areas of GDPR legislation that companies risk violating. One of the key tenets of GDPR is that data requests from users should be explicit and specific. Bundling together your requests with one checkbox is not considered compliant data collection. And the challenge of data collection post-GDPR doesn't just come from the legislation itself; it comes from users, too. According to GWI data from 2024, 34.5% of adult internet users globally now reject cookies at least some of the time.
Attribution In A Privacy-Focused World
The key to accurate attribution is still first-party data. Collecting your own customer data gives you control over compliance and privacy. While there are still gray areas with uncertainty about how GDPR legislation should be interpreted, this will improve as regulators provide more specific guidelines and enforcement increases. Businesses can ensure compliance in the meantime by implementing robust consent mechanisms, providing clear privacy policies and offering easy opt-out options for users.
Once you have that data, the next challenge is using it. Like attribution, data aggregation has always been complex for businesses with smaller marketing budgets. But technology could hold the answer. Data lakes, for example, can make it easier for organizations to store, manage and analyze large, unstructured datasets. They can also ease privacy concerns by anonymizing data for analysis. While this advanced technology still requires time, money and expertise to use effectively today, artificial intelligence is making it more accessible for companies now and in the future.
Machine learning algorithms can also help evaluate converting and nonconverting paths, giving relative value to each and making it easier for marketers to make decisions based on their data. AI can make working with different attribution models easier by combining deterministic data (e.g., logged-in user behavior) with probabilistic models to create a hybrid approach that better estimates cross-device behavior. It can spot patterns that may not have been visible to you before, and with the introduction of agentic models, it can apply insights to adjust budgets in real time and even make decisions on your behalf.
So, while data privacy makes attribution more challenging than ever, it's a welcome challenge for those who value its intentions. Compliant data collection gives users greater control over their data and helps build trust that is sorely lacking in the modern consumer. Combine that with first-party data and new technology, and we can spend marketing budgets more effectively and deliver experiences to consumers that are truly personalized to their behavior, not just based on assumptions.
Forbes Agency Council is an invitation-only community for executives in successful public relations, media strategy, creative and advertising agencies. Do I qualify?
Hashtags

Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles


Forbes
7 hours ago
- Forbes
Don't Let AI Be Your Career Coach
With everyone all tied up in AI-induced knots, worried about their jobs, and concerned about the reliability of AI to begin with, it boggles the mind that so many people have flirted with or worse, committed to AI as a source of career coaching. The come-ons are enticing, creating the impression of infallibility to everything including Kryptonite. There's only one problem, though, That's about as likely to be successful as a robot at a wine tasting. As an independent career and executive coach for the past 28 years, I find the whole thing laughable. Staying up with the times for the sake of progress is one thing; change for change's sake rarely passes muster. Coaching: Linear or Intuitive This conflict between AI algorithms and human wisdom is the crux of the matter. I've coached thousands of people, delivered workshops to tens of thousands, and written articles which have been read by millions. Yet every single person makes a unique impression on me. That's how I learn. To me and countless other coaches, this coaching thing is much more intuitive than you might think. It's a distinctly right-brained process. But is also imprecise. A great coach is an artist Career coaching is more an art than a science; it is more intuitive than it is empirical; it works more on hunches and accumulated experience than on proofs and strictly rational systems of thought. Career coaches who consider themselves scientists are fooling themselves. We are, hopefully, artists – and when we do our jobs well, we produce good work. But it is imprecise. As such, this state of being requires mutual faith – a kind of partnership in belief between the coach and the coached, a shared vision toward which both work. Both must subscribe. Both must be comfortable with the unknown, with ambiguity, with uncertainty. But both must be optimistic, hopeful, and proactive. And while this remains imprecise, it works. Half a century ago, Harold Geneen, CEO and chairman of ITT when it was the largest conglomerate in the world, said, 'Leadership cannot be taught; it can only be learned.' Today, it's the same thing with coaching. Perspective and Experience This thought has given me an uusual perpective. Over the last 28 years, I've coached people from 17 to 82 years old (yes, 82 – really); from the executive suite to the assembly line; from the classroom to the boardroom; from the military and law enforcement to the ex-con looking for a new start; from the private sector to the public and non-profit sectors; from construction and manufacturing to biotech; from science and medicine to sports and leisure; from logistics and transportation to journalism and publishing. I coached people in dozens of industries, with hundreds of job titles and thousands of job descriptions. I've learned from each one, gaining experience and perspective. AI hasn't figured out how to do that. What I've learned as a career coach – more than anything else – is that success almost never comes in one giant leap, but almost always as a result of small steps. It takes perspective and experience – and those traits still belong to us humans alone. Don't let AI be your career coach.


Digital Trends
9 hours ago
- Digital Trends
Meta smart glasses with a built-in display might cost as much as an iPhone
Over the past few years, XR devices have exploded in popularity, and while at it, the costs have also gone up dramatically as the underlying tech keeps pushing new boundaries. For example, Apple's Vision Pro costs $3,500, while the Meta Quest Pro hit the shelves at $1,500. Smart glasses, especially those with a built-in display unit, are also slowly climbing up the price ladder. It seems Meta will buck that trend, or at least beat initial estimates for its next-gen smart glasses that are set to arrive later this year. 'Meta recently figured out a way to slash the price for consumers down to about $800, I'm told. The move stems in part from the company accepting lower margins to boost demand — a common tactic for new products,' says a report by Bloomberg. How do Meta's smart glasses work? Currently in development under the codename 'Hypernova,' Meta had initially planned to hawk the smart glasses at roughly $1,000, while some estimates put the price at $1,400. With the purported $800 asking price, it seems Meta is essentially matching the iPhone 16's sticker value in the market, and possibly, the upcoming iPhone 17, as well. It's pretty obvious that Meta will push these glasses as the next-gen personal computing device, one that is an alternative to smartphones, especially the ubiquitous iPhones in its home market. For comparison, display-equipped smart glasses made by the likes of Xreal and Viture usually fall in the $400-600 bracket, and so do next-gen AI glasses with optical projectors, such as the Even G1. Recommended Videos Meta is essentially pulling off the same formula as Google Glass. Instead of a dual-display system that you will find on smart glasses sold by RayNeo, Viture, and Xreal, Meta's 'Hypernova' smart glasses will only feature a monocular display fitted in the lower portion of the right lens. 'Information will only be displayed in front of the wearer's right eye and will appear most clearly when they are looking downward,' says a Bloomberg report. Powered by Qualcomm silicon, the upcoming Meta smart glasses will feature apps for capturing photos, viewing media, launching maps, and checking notifications. How can they stand out? For more intuitive controls, Meta will reportedly offer a neural wristband that will allow users to control the glasses using wrist gestures and hand movements. Smartwatches such as the Samsung Galaxy Watch 8 have already implemented a gesture-based system for navigating the UI. Notably, the wristband will come bundled in the retail package of the 'Hypernova' smart glasses. Interestingly, the glasses will run a customized version of Android, though there might not be a dedicated app store installed on the wearable. Controls will reportedly be handled by a mix of tap and swipe inputs on the side frame. This is going to be a huge driving force for adoption if Meta and Google can somehow figure out a way to at least access and respond to app notifications coming from your connected phone. But it appears that Meta won't let Google enjoy that cake, especially with Google already working on its own AR glasses built atop the Android XR platform. 'The new version will continue to rely heavily on the Meta View phone app,' reports Bloomberg. The Hypernova smart glasses are expected to arrive in a month from now, and it would be worth waiting to see how they explore AI integration when compared to Google's Gemini on the wearable platform.


Wall Street Journal
9 hours ago
- Wall Street Journal
The Wall Street Journals' News Archive for August 17, 2025
Find what you're looking for Search for topics like "tariffs", your favorite authors, companies or even a more specific query like "dollar's role as a reserve currency".