Flawed Models: Are Brands Losing Millions by Misjudging Affiliate Marketing?

SAN MATEO, CA – May 05, 2026 – A groundbreaking new report suggests that the sophisticated measurement models many of the world’s largest brands rely on to allocate billions in marketing dollars contain critical structural flaws, causing them to systematically undervalue one of digital advertising’s most efficient channels: affiliate marketing.

The research, commissioned by shopping rewards platform Rakuten Rewards and conducted by the data and measurement strategy firm Prohaska Consulting, reveals that standard Marketing Mix Modeling (MMM) is frequently misrepresenting affiliate performance, leading to misguided budget cuts and significant lost revenue. The report, titled "The Next Frontier of Measurement: Fair Evaluation of Affiliates in Marketing Mix Models," draws on dozens of interviews with senior marketing leaders and MMM providers, along with real-world performance data, to expose a costly gap between the channel's actual value and its perceived contribution.

The central finding is a stark warning for CMOs and financial executives: by trusting incomplete data from their MMM frameworks, brands may be actively underfunding a proven channel that connects them with high-intent consumers at the final, crucial moments before a purchase.

The High Cost of Mismeasurement

While over 80% of marketers utilize affiliate programs, the report argues that their true impact is often invisible to the very models designed to measure it. This invisibility has tangible and severe consequences. According to data from Rakuten Rewards, which spans thousands of brands, the affiliate channel delivers an average 17x return in sales versus commissions—a figure that often gets lost in translation.

"Most marketers don't realize how much affiliate performance is being underrepresented in their models, and that gap has real dollar impact," stated Ameet Shah, Partner at Prohaska Consulting, in the report's announcement.

The research presents compelling case studies that put this "real dollar impact" into sharp focus. In one stark example, a major retailer, acting on MMM results that questioned the channel's incremental value, decided to pause its entire affiliate program. The outcome was catastrophic. The brand immediately drove its customers to direct competitors, losing more than half of its prior sales volume from the channel. A year after reactivating the program, the retailer had still not recovered its lost customers or market position—a cautionary tale of the long-term damage that can result from acting on flawed data.

Conversely, the report highlights the immense upside for brands that manage the channel with nuance. When two brands under the same parent company ran affiliate programs concurrently, the one that was actively managed—using strategic tools like elevated Cash Back tiers and limited-time promotions—delivered up to 25 times higher Return on Ad Spend (ROAS) than its passively managed counterpart. Another case showed a leading agency achieving a 10% lift in incremental ROAS simply by properly segmenting its affiliate inputs in planning tools, unlocking value that would have otherwise been missed.

Deconstructing the Measurement Gap

The report identifies two fundamental structural flaws that cause traditional MMM to consistently misread affiliate marketing’s contributions. These are not minor calibration errors but deep-seated biases in how the models are built and fed data.

First is a lack of standardized, granular data. Marketers and their models frequently lump vastly different affiliate activities—such as cash back, coupons, influencer marketing, content partners, and comparison shopping sites—into a single, monolithic "affiliate" bucket. This practice erases crucial performance differences and makes it impossible to discern which specific strategies are driving results. It’s the equivalent of measuring the impact of "video" without distinguishing between a Super Bowl commercial and a 15-second TikTok ad. Furthermore, because the affiliate channel is laser-focused on driving conversions, it often lacks the impression and click data that MMM relies on to measure and compare upper-funnel channels.

The second, more nuanced flaw is that MMM often can't do it alone. Affiliate marketing operates primarily on a commission basis, earning revenue only when a sale is completed. This creates a near-perfect correlation between affiliate activity and sales revenue. To a standard MMM framework, this correlation can be misinterpreted as affiliates simply "claiming credit" for demand that already existed, rather than generating new, or incremental, demand. The model fails to recognize the strategic value of "scale-on-demand"—the unique ability of the affiliate channel to efficiently ramp up sales volume precisely when needed, reaching consumers who are actively making a purchase decision.

A Strategic Reawakening for Affiliates

The implications of these findings extend far beyond the technical realm of data science and analytics, forcing a strategic re-evaluation of the affiliate channel itself. The research positions affiliate marketing not as a simple bottom-of-the-funnel tactic, but as a powerful engine for capturing market share and reaching a valuable audience that is difficult to engage elsewhere.

By misinterpreting the channel's value, brands aren't just misallocating budget; they are creating a vacuum that competitors are all too willing to fill. The report warns that pulling back on affiliates based on incomplete MMM data is a perilous move.

"We've sat across the table from brands that pulled back on affiliates because their MMM told them to do so, and they spent the next year trying to win back ground they didn't need to lose," said Carl Lurie Kalapesi, Chief Commercial Officer at Rakuten Rewards. "Affiliates deliver consistently and at scale, and this research gives marketers evidence and the framework to prove it. Getting the measurement right is more than just a technical fix. It is often the difference between growing market share or handing it to a competitor."

This perspective reframes the conversation from one of pure ROI calculation to one of competitive strategy. The high-intent shoppers who engage with affiliate partners are in the market to buy. If one brand withdraws its presence from these platforms, those shoppers do not simply disappear; they are often driven directly into the arms of a rival.

Charting a Path to Fairer Evaluation

To close this critical measurement gap, the report avoids simply pointing out problems, instead offering a clear and actionable roadmap for marketers, publishers, and the industry at large. The proposed changes aim to elevate the quality of data inputs and the sophistication of the analysis.

For marketers, the recommendations are to:
* Disaggregate the data: Break the affiliate channel into distinct sub-categories within MMM datasets instead of treating it as a single input.
* Track promotional variables: Capture reward intensity and promotional changes, such as shifts in Cash Back rates, as time-series variables to help models understand what drives performance spikes.
* Calibrate with testing: Implement sustainable testing methods like geo-holdouts, time-based tests, and audience-level suppression to prove incrementality and fine-tune MMM results.

For publishers, the path forward involves becoming more active partners in the measurement process by:
* Providing richer data: Supply impression and click data to enable more direct comparisons with other media channels within MMM.
* Building testing capabilities: Develop the infrastructure to support geo-targeting, holdouts, and A/B experiments that help brands quantify their incremental impact.
* Integrating with vendors: Forge direct data integrations with major MMM providers to streamline data sharing and ensure accuracy.

Ultimately, the report calls for a concerted industry-wide effort to establish shared definitions and taxonomies for affiliate sub-categories. This would give the channel the same rigorous measurement footing as established channels like display, video, and native advertising, paving the way for a more accurate and equitable distribution of marketing resources.