As a global leader in SEO, content marketing, and data analytics, Anastasia Braitsik has spent years navigating the complex intersection of consumer behavior and algorithmic shifts. In an era where platform attribution often feels like a hall of mirrors, she has become a vocal advocate for incrementality testing—the rigorous process of determining whether a marketing dollar actually drove a sale that wouldn’t have happened otherwise. Today, we explore her strategic approach to measurement, focusing on how high-level marketers can move beyond “interesting learnings” toward definitive financial impact. We delve into the nuances of prioritizing testing capacity, the role of professional intuition in a data-driven world, and her signature IDEATE framework designed to eliminate post-test bias and drive real-world execution.
With testing capacity being a finite resource in any organization, how do you determine which marketing channels or tactics actually warrant a full incrementality audit versus those that should be left alone?
The reality of modern marketing is that you simply cannot audit every single nuance of your program simultaneously; attempting to do so often leads to a paralysis of analysis. I advise teams to focus their testing capacity exclusively on questions where uncertainty has meaningful financial consequences for the P&L. You have to be incredibly disciplined about asking what you have seen in your observational analyses or Media Mix Modeling that makes you question your current beliefs. If a channel is already well-supported by your sales response and existing data, it doesn’t deserve a testing slot. We prioritize areas where there is a clear discrepancy between what a platform claims and what the business feels, ensuring that every hour spent on measurement has the potential to fundamentally reallocate spend for better returns.
There is often a tension between hard data and a marketer’s intuition; how should a leader lean into their instincts when the numbers on the screen don’t match the broader reality of the business?
There is absolutely a place for instinct in this process, especially when you’ve been close to a program for a long time and something just feels fundamentally off. I often see cases where a platform like Meta ASC reports incredibly attractive marginal CPAs, yet the organization’s overall new customer CAC is consistently getting worse. This sensory friction is exactly why we investigate; it’s possible Meta is finding incremental customers, but it’s just as likely that as we scale, the algorithm is simply taking credit for customers who would have purchased anyway. Trusting that “gut feeling” doesn’t mean you act on it blindly, but rather that you use it as a catalyst to deploy an incrementality test to find the truth. If we find that the channel is indeed highly incremental, we can scale with confidence, but if we’re on the curve of diminishing returns, we have the evidence needed to protect our margins and find a better use for those dollars.
The IDEATE framework emphasizes the “Envision Paths” step before a test even begins; why is it so critical to commit to specific actions before you see the final data readout?
Envisioning your paths before the test starts is the only way to effectively neutralize the human bias that inevitably creeps in once a result is on the table. It is far too easy to sit in a room for 45 minutes and rationalize a poor result by questioning the methodology or dismissing it as a mere “learning” while keeping the budget exactly the same. By writing down exactly what we will do in every scenario—whether it’s an iROAS of 3.0 or 1.5—we remove the emotional weight of cutting spend in a channel that an agency or team has spent years building. For instance, if I have a 50% pre-advertising contribution margin and need a 2.0 incremental ROAS to break even, I need to know today that an iROAS below 1.5 means we are making fundamental changes before another dollar is spent. This level of pre-commitment ensures that the test actually creates value through action, rather than just becoming another slide in a deck that everyone ignores.
Once an incrementality test is completed, how can marketers ensure that the insights don’t just gather dust but instead serve as a living reference point for daily operations?
The true power of an incrementality test lies in its ability to provide a “calibration filter” for your everyday platform metrics until the next major audit. If Meta reports a ROAS of 4.0 during a test, but our incrementality measurement shows an iROAS of only 2.0, we now know that the platform’s reported return is twice the actual incremental return. We shouldn’t assume this ratio is a permanent law of nature, but it serves as a critical working benchmark for managing the channel in the weeks that follow. If that platform ROAS eventually dips to 3.5, I can quickly calculate that our real iROAS might have fallen to around 1.75, which means those dollars are no longer paying for themselves given our 50% margin. This gives the team a clear, math-based reason to act—either to optimize, reduce spend, or pivot the strategy—without having to wait for a brand-new, multi-week study to conclude.
As we look toward scaling sophisticated measurement programs, how does the final stage of “Executing on Findings” and documenting the loop transform a single test into a long-term competitive advantage?
Finishing the loop is about more than just moving a budget slider; it’s about building an organizational memory that prevents teams from relearning the same painful lessons every six months. By documenting the results and the subsequent actions in one central repository, we give every future test a much more informed starting point. The IDEATE process—Insight, Draft Hypothesis, Envision Paths, Arrange, Track, and Execute—only works if the “Execute” phase is taken as seriously as the “Insight” phase. When you prioritize questions with material P&L impact and follow through on pre-agreed actions, you may find yourself running fewer tests overall, but the ones you do run will have a significantly higher probability of fundamentally changing the trajectory of the business. It turns measurement from a defensive reporting chore into a proactive offensive weapon that ensures every dollar is working as hard as possible.
What is your forecast for marketing measurement?
I believe we are entering a period of radical transparency where the “black box” of platform attribution will finally lose its grip on the executive suite. As we move deeper into this year, the reliance on real-time experimentation and incrementality will shift from being a “nice-to-have” for sophisticated brands to a survival requirement for any business with tight margins. We will see a significant move away from vanity metrics and toward integrated systems that blend Media Mix Modeling with frequent, tactical lift studies. Marketers who can master the art of “Envisioning Paths” and acting on data with cold, mathematical precision will outperform those who remain trapped in the cycle of chasing platform-reported ROAS. The future belongs to those who view measurement not as a way to prove they were right, but as a rigorous system for making sure they aren’t accidentally wasting half their budget on customers they already had.
