Modern media measurement faces a significant hurdle, especially when dealing with large-scale, multi-channel investments like TV and Connected TV (CTV). Marketers frequently struggle with inaccurate or inflated Return on Investment (ROI) reporting, making it difficult to pinpoint the true value of their ad spend.
Establishing a robust pre-campaign baseline is the foundational step needed to solve this problem and accurately calculate true incremental lift. This disciplined approach ensures that media decisions are grounded in actual cause-and-effect data, rather than potentially misleading attribution figures. Keep reading to learn how to establish a comprehensive baseline performance and use it to measure true incrementality.
The Fundamental Problem with ‘Observed’ Performance Metrics
Performance metrics often rely on "observed" performance, which tracks all conversions seen after a user has been exposed to an advertisement. While this data looks compelling on paper, observed lift frequently overstates success because it attributes success to the campaign even when the user would've converted naturally. The ultimate goal of sophisticated media is to drive incremental lift, which is the direct, causal impact the advertisement has on sales that wouldn't have otherwise happened.
Consider a customer who regularly buys a product every six weeks. If they see a TV ad and then convert during that six-week window, standard attribution models often credit the ad with the sale, even though the customer was already primed to purchase. For example, a fitness apparel company spending $100,000 monthly on Meta ads received a platform-reported Return on Ad Spend (ROAS) of 4:1. They later discovered their true incremental ROAS was only 2.5:1, proving that 37% of the attributed revenue would have occurred without the ads. This flawed approach fails to isolate the true value that the new media spend adds to the business.
Attribution Flaws in Non-Clickable Media
Traditional attribution models, such as last-click or simple multi-touch attribution, are particularly insufficient when measuring upper-funnel media channels. Channels like linear TV, radio, and Connected TV (CTV) are non-clickable or involve highly delayed, indirect conversion paths. This is a critical problem because nearly 70% of the US population now uses Connected TV.
Attempting to force a click-based model onto these broad channels naturally leads to missed conversions or wildly inaccurate reporting. To accurately prove the value of a major media investment, marketers must move beyond click reliance and toward true causality. The necessary method compares the performance of an exposed audience against a statistically similar unexposed audience, regardless of whether a click occurred.
Some YouTube campaigns, for instance, have shown platform metrics underreported the campaign's true impact by 70%, meaning actual incremental ROAS was 3.4 times higher than reported figures. This comparative methodology is the only reliable way to prove causal impact and understand the true value these channels truly deliver.
The Systematic Approach to Establishing Your Baseline
Establishing a solid baseline performance is the first systematic step toward accurate measurement. A robust baseline requires a consistent historical data set, typically spanning three to six months, to account for cyclical patterns and external factors. This historical analysis is important for identifying meaningful patterns in performance metrics, such as conversion rate fluctuations by day of the week or seasonal engagement spikes.
The baseline documents what would've happened anyway without the new media spend, which is the attributed value of everything outside the campaign being measured. The focus during this phase is on consistent data collection before any new media goes live.
Identifying Critical Pre-Campaign KPIs (Key Performance Indicators)
Choosing the right metrics for the baseline requires shifting focus away from easily manipulated vanity metrics toward KPIs that directly impact business results. The baseline should center on holistic and direct-site performance metrics that upper-funnel media is intended to influence. Critical pre-campaign metrics include the overall website conversion rate, the average Customer Acquisition Cost (CAC), and the volume of unexposed or direct sales.
It’s also important to track both leading and lagging indicators during the baseline period. Lagging indicators, such as final sales volume and revenue, provide the ultimate picture of success. Leading indicators, like spikes in website traffic or organic search volume related to the brand, offer early signals of consumer intent. Specifically, tracking Branded Search Volume provides a measure of awareness driven by upper-funnel media before a transaction occurs.
Only one-third of Chief Marketing Officers (CMOs) regard the ROI of total marketing spend as a strategic KPI, revealing a critical opportunity for baseline establishment to bridge the measurement gap. By focusing on these comprehensive business-level metrics, the team ensures the campaign’s success is measured against actual business growth, not just platform statistics.
Documenting External & Market Conditions
A baseline is fundamentally incomplete if it doesn't account for the external factors that influence organic performance, regardless of the new media campaign. Documenting non-marketing-related variables and broader market-level data in the pre-launch phase helps isolate the campaign's true effect later. These external conditions provide the necessary context to interpret performance fluctuations accurately once the campaign is underway.
Seasonality is a key factor that must be precisely documented. This involves tracking known seasonal highs and lows and understanding historical trends for the product or industry. While seasonality is a real and powerful force in business, it mustn't become a convenient scapegoat that masks deeper, more controllable issues within marketing or user experience.
Furthermore, documentation must cover competitor activity and macroeconomic trends. This includes noting any recent or planned major media campaigns, product launches, or significant news from key competitors. Broader market shifts, such as policy changes or major holiday periods, should also be documented, as these events can independently cause spikes or dips in baseline performance.
Designing the Incrementality Test: Isolating Causal Lift
Once the baseline provides a clear picture of what would happen anyway, marketers use this data as the control benchmark to design rigorous incrementality tests. This process scientifically isolates the causal impact of the new media spend, moving beyond simple attribution figures.
Incrementality testing is a scientific method, similar to A/B testing, designed to isolate the campaign's true, causal impact. It determines the value of advertising by comparing the behavior of a test group that sees the ads against a statistically identical control group that does not.
Test Methodologies for National Campaigns
Large-scale media buyers must utilize sophisticated methodologies to effectively isolate media exposure across broad channels like TV and CTV. One highly effective approach for national campaigns is Geo-Testing, which splits a target audience into statistically equivalent geographical test markets and control markets. This powerful method allows marketers to reliably compare sales performance across regions where the campaign ran versus regions where it did not, isolating regional differences. Working with a dedicated media agency for geo-testing national campaigns ensures statistical rigor.
Another useful method is Public Service Announcement (PSA) or "Dummy Ad" testing. This involves exposing the control group to a non-campaign, generic advertisement on the same channels and time slots used by the brand's creative. This accounts for the general channel-level media effect, such as the overall attention generated by being on TV, without allowing the brand's specific creative or message to influence the control group.
Isolating Impact with Advanced Holdout Strategies
While models like multi-touch attribution (MTA) struggle with upper-funnel, non-clickable placements, incrementality testing proves far more accurate for platforms like YouTube and CTV. Advanced media measurement solutions for large brands often integrate platform-specific measurement tools, like Conversion Lift experiments, into broader cross-channel geo-testing frameworks. This level of rigor helps move past the limitations of simple MTA vs incrementality testing debates.
Holdout Groups are often implemented by randomly assigning a percentage of the audience within the target demographic not to be served the advertisement. This logistical complexity requires precise audience management but offers highly specific data on the direct causal lift among the audience the brand cares about most. These methods require significant statistical rigor to ensure the test and control groups truly are identical before the campaign starts. Businesses now have better access to these capabilities, as Google reduced the minimum spend requirement for incrementality experiments from upwards of $100,000 to just $5,000.
Translating Baseline Data into a Test Hypothesis
The comprehensive data gathered during the baseline period is important for formulating a clear, testable hypothesis for the incrementality test. This hypothesis should establish the expected result, such as "The new CTV campaign will drive X percent incremental conversion lift above the established Y baseline."
To make the test meaningful and actionable, the team must first set a minimum detectable effect. Defining the minimum detectable effect means specifying the smallest lift that would still be statistically significant enough to justify the entire media spend.
Marketers also use the baseline period to calculate historical standard deviation, which is required for the test's statistical power calculation. Critically, the control group criteria must be specified, emphasizing that the control group must be statistically identical to the test group in all relevant aspects, including demographic profile and historical purchase frequency.
Proving Value: Reporting True Incrementality to Leadership
The ultimate goal of rigorously establishing a baseline and performing incrementality testing is to furnish executive leadership with unassailable data on Return on Investment (ROI). This structured approach prevents the common pitfall of reporting inflated or misleading figures that merely capture existing demand. Reporting true incremental lift builds trust in the marketing function and enables much smarter, data-driven budget allocation decisions.
Calculating True Incremental ROI
Calculating true incremental ROI involves a straightforward comparison built on the test and control group data. The key difference is the number of conversions driven solely by the advertising exposure. The formula for identifying these conversions is: Incremental Conversions = Exposed Group Conversions - Control Group Conversions.
Once the incremental conversions and their associated revenue are calculated, the True Incremental ROI formula is applied. This formula is: (Incremental Revenue - Campaign Cost) / Campaign Cost. This clear, causal link allows leadership to understand exactly how much the new media channel is genuinely adding to the bottom line, rather than just capturing existing demand.
Sophisticated advertisers strive to calculate incremental ROAS accurately for all media channels, including Connected TV, to ensure their reports are based on true causal impact, not observed results.
Optimizing Budget Allocation with Baseline Insights
The value of the baseline and incrementality data extends far beyond the initial reporting phase; it provides the insight needed to optimize future media buys effectively. This data allows for strategic budget redirection based on where the campaign is proven to drive new conversions. Marketers can now move budget away from audiences or channels that showed a high "baseline conversion rate," meaning those people were likely to convert anyway.
Instead, the budget is strategically allocated toward the audiences and placements where the campaign proved it could drive genuine incremental lift. This level of optimization maximizes efficiency and ensures a higher percentage of the budget is dedicated to true growth. Furthermore, 80% of US senior marketing analytics professionals reported that implementing insights derived from incremental experiments has a high impact on revenue growth.
Drive Proven Growth with Advanced Media Measurement
Establishing a robust pre-campaign baseline and applying rigorous incrementality testing is the only reliable way to accurately measure the impact of large-scale, national media campaigns. This disciplined approach eliminates the ambiguity of observed performance and replaces it with clear, causal data on how your media spend is driving new customer acquisition. True growth relies on knowing what your investment is genuinely adding to the bottom line, not just what it touches.
We understand the complexities of sophisticated media buying across channels like TV, YouTube, and Connected TV. Our approach integrates baseline performance analysis and meticulous incrementality testing from the planning stage onward, ensuring every dollar spent delivers provable value.
We're dedicated to helping large brands achieve validated, incremental return on their media investments. Contact us today for more information, and let us develop a media buying and campaign strategy built on precision and proven impact.
