By [Author Name]
Published: September 2026
Executive Summary: The Core Dilemma of Modern Marketing
In the high-stakes world of digital commerce, Return on Ad Spend (ROAS) reigns supreme. Advertisers, CMOs, and finance departments treat the metric as the ultimate litmus test for campaign viability. It is simple, mathematically straightforward, and easy to dashboard. However, according to industry experts, ROAS is dangerously flawed when viewed in isolation.
Without accurate, nuance-aware attribution, ROAS can easily morph into a vanity metric that misleads companies into overfunding underperforming channels while starving truly incremental growth drivers. This systemic blind spot is prompting a hard look at how digital marketers calculate success, particularly within fast-growing sectors like retail media networks (RMNs).
Main Facts: The Anatomy and Achilles’ Heel of ROAS
At its core, ROAS serves a vital function: it connects advertising investment directly to generated revenue. The calculation is deceptively simple:
$$textROAS = fractextSales Attributed to AdstextCost of Ads$$
Armed with this metric, marketers compare competing campaigns, justify budgets to executive leadership, and allocate capital dynamically.
Yet, the metric possesses a profound Achilles’ heel. Mike Murphy, Vice President of Marketing at attribution firm Incremental, warns that ROAS becomes actively misleading when attribution models grant ads too much or too little credit, or when they completely ignore underlying purchase intent and organic baseline sales.
The Last-Touch Fallacy in Retail Media
Consider a standard digital commerce scenario: An ecommerce brand allocates $10,000 to a retail media campaign on a major platform. Using a traditional last-touch attribution model, the marketing dashboard reports $50,000 in attributable sales tied directly to those ads. The resulting ROAS sits at an enviable 5:1 ratio—meaning every single dollar invested returns five dollars in revenue.
To an executive team operating purely on dashboard metrics, the decision is obvious: pump more money into the channel.
However, this calculation ignores a critical nuance. "ROAS credits the last ad touchpoint before a sale, whether or not it caused anything," Murphy explains. "That leaves a very big gap. ROAS can take credit for organic sales that would have happened anyway, or for a sale an earlier touchpoint drove. It also only sees what can be tracked directly."
In environments like Amazon, Walmart Connect, or specialized retail media networks, consumer buying intent is already extraordinarily high. Shoppers are actively browsing with credit cards in hand; they are not passive social media scrollers discovering a brand for the first time. Consequently, last-touch attribution frequently rewards brands for sales they would have captured organically anyway.
Chronology & Evolution: How Attribution Got Broken
To understand how digital marketing arrived at this precarious juncture, it is helpful to trace the evolution of ad measurement over the past two decades.
Phase 1: The Wild West of Direct Response (Early 2000s)
In the early days of ecommerce, tracking was rudimentary. Marketers relied heavily on "last-click" attribution because it was the only data reliably captured by pixel technology. Platforms graded their own homework, leading to inflated claims of effectiveness that largely went unchallenged.
Phase 2: Multi-Touch Attribution and Privacy Crackdowns (2010s–2020s)
As customer journeys grew more complex—spanning mobile apps, desktop searches, social media, and marketplaces—the industry shifted toward Multi-Touch Attribution (MTA). Brands attempted to distribute credit across every touchpoint a consumer interacted with prior to purchase.
However, this progress was severely disrupted by sweeping privacy regulations (such as Apple’s App Tracking Transparency framework) and the deprecation of third-party cookies. As deterministic tracking degraded, platforms pivoted heavily toward modeled data and walled-garden ecosystems, particularly retail media networks.
Phase 3: The Retail Media Boom and the Incrementality Awakening (Present Day)
Today, retail media represents the fastest-growing sector of digital advertising. Because RMNs sit directly on the point of purchase, they offer rich closed-loop reporting. Yet, this proximity has created a false sense of security. Advertisers assume that because a sale happened on the same platform as the ad, the ad caused the sale. Industry leaders are now pushing past traditional attribution models to embrace "incrementality testing"—seeking to answer the fundamental question: Would this sale have occurred if the ad had never run?
Supporting Data: Unpacking Incrementality and Cross-Device Blind Spots
The gap between standard attribution and true incrementality can fundamentally alter a brand’s unit economics.
The Organic Overlap Effect
Returning to the earlier scenario of the $10,000 retail media spend: suppose the brand’s analytics team digs deeper into the data. They discover that roughly 50% of the time the retail media ad appears, the brand’s products are already naturally visible near the top of the organic search results page.
- Scenario A (No Organic Visibility): When the product is invisible organically, the ad deserves full credit for capturing the shopper’s attention and driving the conversion.
- Scenario B (Organic Dominance): When the ad appears directly alongside a top-ranking organic result, the ad is essentially paying for traffic the brand would have secured for free.
When factoring in true incrementality—distinguishing between forced conversions and organic cannibalization—the effective ROAS drops from a misleading 5:1 down to a realistic 3:1.
The Inverse Problem: Undervaluing Ads
Conversely, the attribution problem can work in reverse, causing ROAS to register artificially low. An ad may successfully generate consumer demand and drive revenue that a standard attribution window or pixel completely fails to capture.
For instance, a shopper might encounter a sponsored product ad while browsing on a mobile device within a retail marketplace, but later complete the transaction by navigating directly to the merchant’s independent direct-to-consumer (DTC) website on a desktop computer.
In fragmented buyer journeys, cross-device and cross-channel blind spots mean the initial ad sparked the purchase, but the attribution system remains entirely blind to it.
To combat this, advertising giants like Google utilize complex conversion modeling—employing machine learning to estimate conversions across devices and privacy restrictions. Google explicitly notes that without such modeling, reported metrics would capture only a fraction of true campaign performance. If an ad drives a sale that attribution fails to recognize, the calculated ROAS is deceptively low, risking the premature termination of a highly profitable campaign.
Official Responses and Industry Perspectives
As the limitations of conventional ROAS become impossible to ignore, marketing executives and data scientists are redefining how performance is measured.
Independent auditing firms and attribution specialists are sounding the alarm against dashboard complacency. According to Mike Murphy and fellow industry analysts, brands must stop viewing ad-platform dashboards as objective sources of absolute truth. Because ad networks have a vested financial interest in proving high performance to retain budgets, self-reported metrics are inherently biased toward over-attribution.
Furthermore, privacy-first legislation enacted by global regulators has effectively dismantled deterministic tracking. Industry bodies now emphasize that brands must pivot from asking "What did my ads touch?" to "What incremental profit did my ad spend actually unlock?"
Implications: Building a Better Measurement Framework
If traditional ROAS is flawed, how should modern ecommerce brands evaluate advertising efficacy? Experts recommend a multi-layered approach centered on testing, holistic metrics, and financial reality checks.
1. Execute Incrementality Testing
Marketers must actively test whether attributed sales are genuinely incremental.
- For Smaller Budgets: Mike Murphy suggests simple holdout or geo-matched tests. A company can pause advertising on a specific group of products for several weeks while maintaining baseline spend on a control group, then compare the net change in total sales. While directional rather than hyper-precise, these tests reveal the true macro-impact of the ad spend.
- For Enterprise Budgets: Larger advertisers with deep relationships with retail media networks should leverage randomized control trials (RCTs) and advanced geographic testing to isolate ad lift from organic momentum.
2. Look Beyond ROAS to Contextual Metrics
Relying on a single metric invites strategic blind spots. Brands should evaluate campaign health through a broader lens, incorporating metrics such as:
- Customer Acquisition Cost (CAC) Payback Period: How long does it take for gross margin from a new customer to cover the cost of acquiring them?
- Blended ROAS: Total company revenue divided by total marketing spend across all channels, which smooths out channel-specific attribution quirks.
- New-to-Brand (NTB) Percentage: Measuring whether campaigns are genuinely expanding market share by capturing new customers rather than remarketing to loyal brand advocates.
3. The P&L as the Ultimate Source of Truth
Ultimately, dashboard metrics are approximations; corporate financial statements are reality. If a marketing team scales ad spend upward by 50%, the company’s overall top-line revenue, gross profit, and bottom-line net income should scale commensurately. Conversely, if ad spend is slashed, net profitability should experience a measurable contraction.
"Your P&L should be your first source of truth—it doesn’t lie," Murphy emphasizes.
Conclusion
Return on ad spend remains an indispensable compass for digital marketers, but it can no longer be treated as an infallible oracle. By acknowledging the limits of last-touch attribution, accounting for organic baseline cannibalization, and validating channel metrics against bottom-line profitability, ecommerce brands can move past misleading dashboards and build resilient, genuinely profitable growth strategies.
