How to market your business in light of iOS privacy

In the world of digital marketing, privacy changes are like sudden weather shifts during a mountain climb. The path you’ve been following suddenly becomes obscured, familiar landmarks disappear in the fog, and you need to quickly adapt your navigation strategy. Just as experienced climbers rely on multiple tools beyond just visual cues—compasses, altimeters, and topographic maps—today’s marketers need to diversify their measurement approaches beyond user tracking.

Apple’s iOS data privacy changes represent perhaps the most significant “weather event” in recent digital marketing history, forcing brands to reconsider how they track, measure, and optimize their marketing efforts. But this challenge isn’t insurmountable—it’s an opportunity to build more robust, future-proof marketing strategies.

The current state of privacy regulation and updates

The data protection and digital privacy landscape continues to evolve rapidly, with significant implications for marketers. Understanding these changes and how they change targeted ads is the first step toward adapting your marketing strategy effectively.

Apple’s App Tracking Transparency (ATT) framework, introduced on Apple devices with iOS 14.5 in 2021, marked a pivotal shift by requiring explicit user consent for tracking across apps and websites. More recent iOS 17 updates have further enhanced privacy protections with features like Mail Privacy Protection, which prevents senders from knowing when users open emails and masks IP addresses.

Meanwhile, Google has been planning to phase out third-party cookies in Chrome, though the timeline has faced multiple delays. Despite these postponements, the writing is on the wall—cookie-based tracking is living on borrowed time.

Beyond these platform changes, frameworks for data privacy regulation continue to expand globally:

• The EU’s GDPR (General Data Protection Regulation) remains the gold standard for privacy regulation, imposing strict requirements on data collection and processing
• California’s CCPA and CPRA have established strong privacy protections in the U.S.’s largest state economy
• New state-level privacy laws continue to emerge across Virginia, Colorado, Connecticut, and beyond
• Countries like Brazil (LGPD) and China (PIPL) have implemented their own comprehensive privacy frameworks

Recent statistics show that when given a clear choice, approximately 80% of iOS users opt out of tracking on their Apple devices, severely limiting the data available to advertisers through traditional means. This high opt-out rate underscores why relying solely on user-level tracking has become increasingly untenable.

Marketing your brand with increased privacy restrictions

It’s crucial to understand that privacy changes like Apple’s App Tracking Transparency framework don’t fundamentally alter which marketing channels work for your brand. Facebook, Google, TikTok, email marketing, and other channels that worked before iOS privacy changes will likely continue to be effective, even if targeted advertising is no longer what it once was. What’s changing isn’t the path itself, but rather your ability to see it clearly.

To return to our mountain climbing metaphor, the trails haven’t disappeared—they’re just harder to see through the fog. The mountain itself hasn’t changed shape; your visibility has. The most effective routes to reach your customers still exist, but your instruments for measuring progress and effectiveness need to adapt.

This distinction is important because some marketers mistakenly believe they need to completely overhaul their marketing strategy in response to privacy changes. The “State of Privacy in Advertising” study found that 88% of marketers polled believed data privacy laws would moderately to significantly impact their ability to deliver personalized advertising. In reality, the bigger shift needs to happen in how you measure, attribute, and optimize your marketing efforts, not necessarily in the channels or tactics themselves. (In the study, only 23% of marketers anticipated challenges in measurement and validation, while only 6% expected issues with planning and forecasting.)

For most brands, this means:

• Continuing to leverage the channels that have historically driven results
• Adapting the measurement approaches used to evaluate those channels
• Developing new frameworks for testing and optimization that don’t rely on perfect user-level data
• Using probabilistic forecasting to plan effectively without the need for user tracking

The marketing landscape hasn’t fundamentally changed—consumers still use the same platforms and respond to compelling messaging, even if you don’t have the third-party data to see it. What’s changed is your visibility into exactly how and when those interactions lead to conversions. Acknowledging this reality allows you to focus on adapting your measurement approach rather than abandoning effective marketing channels out of measurement frustration.

How privacy changes impact different marketing measurement approaches

With privacy regulations tightening and increased data security, different measurement methodologies face varying degrees of challenge. Understanding these limitations can help you determine which approaches remain viable in your marketing stack.

Multi-touch attribution (MTA)

Multi-touch attribution models attempt to track individual users across touchpoints to assign credit for conversions. Unfortunately, this methodology has been hit hardest by privacy changes.

MTA relies heavily on consistent user identification across platforms and touchpoints—precisely what ATT and cookie deprecation disrupt. When a user opts out of tracking on their Apple account or blocks cookies, they essentially disappear from view in the middle of their journey, creating significant blind spots in your attribution model.

The immediate effects include:

• Fragmented user journeys where the connection between ad exposure and conversion is broken
• Inability to accurately attribute value to upper-funnel marketing activities that don’t lead directly to conversion
• Growing discrepancies between MTA models and actual business results

Many marketers still using MTA find themselves making decisions based on increasingly incomplete data. That can lead to misallocated budgets and missed opportunities or focusing on the lowest parts of their funnel, missing out on crucial growth strategies.

Incrementality testing

Incrementality testing aims to measure the true impact of advertising by comparing test groups (exposed to ads) to control groups (not exposed). We have some issues with this approach (you can read about those more in some of our other pieces like our explainer on why incrementality tests are not rigorous randomized controlled trials (RCTs), which is coming soon), but it also faces mounting challenges in the privacy-first era.

Creating clean test and control groups has become substantially more difficult as:

• User identification becomes less reliable, making it harder to maintain group integrity
• Geo-testing (using different geographic regions for test and control) faces challenges with regional variations that can skew results
• Cross-device behavior becomes nearly impossible to track accurately
• The cost of running statistically significant tests continues to rise due to the need for larger sample sizes to overcome data limitations

Additionally, incrementality tests typically capture only a specific moment in time, making it difficult to understand how marketing performance changes over extended periods or during different seasons. This point-in-time limitation can lead to misleading conclusions when extrapolated to overall marketing strategy.

While incrementality testing can provide valuable insights, its growing complexity and cost make it less practical as a standalone measurement solution.

Platform-reported data

Ad platforms like Meta, Google, and TikTok have responded to privacy changes by implementing various modeling approaches to estimate conversions they can no longer directly observe. While better than nothing, these approaches create their own challenges:

• Each platform uses different methodologies for modeling conversions, leading to inconsistent measurement standards
• Platforms inherently have incentives to show positive performance, potentially inflating their contribution. (They’re “grading their own homework.”)
• Modeled conversions often don’t align with actual business results, creating a “reality gap”
• Cross-platform attribution remains problematic, with multiple platforms potentially claiming credit for the same conversion

Many marketers report seeing platform-reported ROAS that, when summed across channels, exceeds their total business revenue—a mathematical impossibility that highlights the reliability issues with platform data.

While major platforms like Meta and Google are investing heavily in developing privacy-compliant measurement alternatives such as Meta’s Conversion API and Google’s Enhanced Conversions, these solutions face fundamental limitations. These platform-specific fixes attempt to model what they can no longer directly observe, but they’re doing so without the benefit of pixel-based or third-party data that previously powered their reporting. This creates two critical problems for marketers:

• Each platform’s solution operates in isolation, leading to fragmented measurement across your marketing ecosystem
• These solutions inevitably contain platform bias, as they’re designed to demonstrate their own effectiveness

This makes it imperative for brands to implement attribution solutions that operate independently from platforms using statistical methods. Unlike platform-specific fixes that will continue to evolve (and potentially disrupt your measurement) with each privacy update, statistical approaches like Prescient’s MMM are inherently adaptable. They validate performance through multiple data sources rather than relying on increasingly limited tracking technologies, giving you a consistent, unbiased view of your marketing performance regardless of ongoing platform changes.

Emerging platforms: testing viability in a privacy-first world

As established channels become more competitive and expensive, marketers are increasingly exploring newer advertising platforms like AppLovin, Digital Turbine, and Connected TV (CTV). Testing these platforms presents unique challenges as we increase data protection frameworks.

The fundamental question becomes: how can you determine if these new channels deliver meaningful ROI when traditional measurement approaches are compromised?

This is where statistical approaches prove particularly valuable. Instead of relying on direct user tracking, statistical methods look for correlations between marketing activities and business outcomes at an aggregate level. This approach allows marketers to:

• Run smaller proof-of-concept campaigns that can detect signal without requiring massive budget commitments
• Understand how new channels interact with existing marketing efforts through holistic modeling
• Make data-driven decisions about scaling investment without perfect attribution

For example, CTV advertising has been notoriously difficult to measure through traditional attribution methods due to its walled-garden nature and the disconnect between viewing devices and conversion devices. Statistical modeling can help bridge this gap by identifying the relationship between CTV campaign flights and changes in website traffic, branded search volume, and conversion rates—even without direct user tracking.

The case for statistical modeling in privacy-first marketing

As user-level tracking becomes increasingly unreliable, marketing mix modeling (MMM) has experienced a renaissance. Unlike approaches that depend on tracking individuals, MMM uses aggregate data and statistical techniques to understand the relationship between marketing inputs and business outcomes. But that’s not to say that all MMMs on the market are the same or equally equipped to help marketers navigate this increase in data protection.

Modern MMM solutions like Prescient AI have evolved far beyond the quarterly, channel-level models of the past. MMM offerings and abilities differ, but Prescient’s approach offers:

• Campaign-level insights that provide actionable granularity without requiring individual user data
• Daily model updates that support agile decision-making
• The ability to capture complex interactions between channels, including halo effects
• Future-proof measurement that doesn’t rely on cookies, device IDs, or user tracking
• Reveals how campaigns contribute revenue to each sales channel for omnichannel brands
• Predicts profitability of future spend allocation

This statistical approach works by analyzing patterns in historical data to understand how different marketing activities correlate with business outcomes. By controlling for external factors like seasonality and reflecting the realities of the marketing environment, Prescient’s MMM can help you understand the true impact of your marketing efforts and plan your optimal spend.

Most importantly, MMM doesn’t require personally identifiable information or user tracking to function effectively. It relies instead on aggregate spend data, impression volumes, and business metrics that aren’t compromised by privacy regulations. This makes it inherently more stable and reliable in today’s changing privacy landscape.

Creating a comprehensive measurement strategy

Rather than viewing different measurement approaches as mutually exclusive, forward-thinking marketers are developing comprehensive measurement strategies that leverage the strengths of multiple methodologies while acknowledging their limitations.

A robust measurement strategy in today’s privacy landscape might include:

1. Marketing mix modeling as the foundational measurement approach, providing a holistic view of marketing effectiveness across channels and campaigns
2. Platform-reported data as a directional signal for optimization within platforms, with the understanding that it may not perfectly reflect true business impact
3. Incrementality testing for specific strategic questions or to validate findings from other measurement approaches
4. First-party data analysis to understand customer behavior post-acquisition by collecting data from the customers willing to volunteer these insights

The key is to understand what each approach can reliably tell you and where its blind spots lie. For example, while platform data might help optimize creative selection within Meta campaigns, MMM provides a more reliable view of Meta’s overall contribution to your business compared to other channels.

This multi-layered approach provides redundancy and cross-validation, reducing the risk of making decisions based on incomplete or misleading data from any single measurement source.

Practical steps for privacy-first marketing

The most successful brands will be those that view privacy changes not as obstacles but as opportunities to develop more sophisticated, consumer-respectful marketing approaches. Beyond measurement, there are several practical steps marketers can take to adapt to advanced data protection:

Strengthen first-party data collection

With third-party data becoming less reliable, your owned customer data becomes increasingly valuable. Consider:

• Implementing a thoughtful strategy for email collection and progressive profiling
• Creating compelling reasons for customers to authenticate and share information directly
• Building a robust customer data platform (CDP) to unify first-party data across touchpoints
• Developing segmentation strategies that don’t rely on individual tracking

First-party customer data can be helpful not only with measurement but also effective targeting through platforms’ custom audience capabilities, even as third-party targeting options diminish. As long as you’re leveraging robust data handling practices, this information can help you continue to deliver personalized ads. Do keep in mind, however, that first party data sourcing can be tricky; methods like post-purchase surveys can be rife with inaccurate self-reported data.

Shift from user-level to cohort-based optimization

Rather than chasing individual user optimization, focus on identifying patterns at the cohort level:

• Analyze performance by creative themes, messaging approaches, and audience segments
• Optimize for early indicators of success (like engagement rates) that can be measured without conversion tracking
• Use these insights to inform your targeting and creative strategy even when perfect attribution isn’t possible

This approach aligns well with platforms’ own shift toward privacy-preserving technologies like Google’s Topics API and Meta’s Aggregated Event Measurement.

Adopt a test-and-learn mindset

With perfect measurement no longer possible, structured experimentation becomes even more important:

• Implement a systematic testing program for creative, audiences, and channels
• Consider incrementality tests for answering specific, point-in-time questions
• Leverage a platform using statistical modeling to identify the signal amid the noise
• Uncover ignored optimizations to carve out a budget for smaller experiments

Testing newer platforms like AppLovin or CTV requires this mindset, starting with smaller tests that can be measured through statistical approaches before scaling investment.

Wrapping it up…

The data privacy landscape has fundamentally changed how marketers must approach measurement and optimization. Rather than fighting these changes or hoping for a return to the tracking-heavy past, forward-thinking marketers are adapting their strategies to thrive in this new environment.

Statistical approaches like the advanced marketing mix modeling offered by Prescient are a compelling path forward, providing reliable insights without depending on increasingly restricted user-level tracking. By combining these approaches with strategic use of platform data and other data sources that respect data privacy restrictions, marketers can develop a comprehensive view of performance that drives better business decisions.

Ultimately, the marketers who will succeed in this data privacy-first world are those who:

• Adapt their measurement approaches to the new reality
• Focus on the metrics that truly matter to business outcomes
• Test emerging platforms thoughtfully with appropriate measurement
• View privacy as an opportunity to build more sustainable marketing strategies

At Prescient AI, we’re helping brands navigate this transition with our advanced marketing mix modeling platform that provides campaign-level insights without relying on cookies or user tracking. Our approach is future-proof by design, giving marketers the confidence to make decisions even as the data privacy landscape continues to evolve.

Ready to explore how statistical modeling can strengthen your marketing measurement in a privacy-first world? Request a demo today.