Former Meta employee alleges artificial ROAS inflation for Shops ads

A former Meta product manager has filed a whistleblower complaint alleging the social media company artificially inflated return on ad spend metrics for its Shops ads product by counting shipping fees and taxes as revenue, subsidizing auction bids, and failing to disclose calculation differences to advertisers.

Samujjal Purkayastha filed the complaint with the London Central Employment Tribunal on Wednesday, August 20, according to documents viewed by ADWEEK. The former employee worked on Meta's Shops ads team from March 2022 until his termination on February 19, 2025.

According to the complaint, Meta data scientists discovered during internal reviews in early 2024 that Shops ads return on ad spend had been inflated between 17% and 19%. The discrepancy resulted from Meta including shipping fees and taxes in sales calculations, even though merchants never received that money, Purkayastha alleged.

Meta's standard advertising products excluded these figures, following the same approach as competitors like Google. Without shipping fees and taxes, Shops ads performed no better than Meta's traditional advertising campaigns, the complaint stated.

"This was significant," the filing reads. "In addition to the ROAS performance metric being overstated by nearly a fifth, it meant that, rather than having exceeded our primary target, the Shops Ads team had in fact missed it once the figure was reduced to take account of the artificial inflation."

Calculation discrepancies masked performance differences

Around mid-2022, Purkayastha discovered Meta calculated ROAS for Shops ads using conversion values that included shipping fees and taxes in the sale price. Standard Meta ads and rival platforms used "net" figures excluding those fees when reporting return on ad spend to buyers.

Advertisers had limited visibility into this discrepancy because Meta's Ads Manager dashboard displayed performance without disclosing how Shops ads ROAS was calculated. The data suggested Shops ads outperformed standard "business as usual" campaigns that direct users outside Meta's applications, with internal reports showing 70% to 80% of Shops ads exceeding BAU ads performance, well above the company's 60% benchmark.

Meta declined to respond to requests for comment at the time of publication.

The alleged inflation occurred as Meta sought to recover from Apple's 2021 privacy changes that significantly reduced iOS tracking data. Meta's former chief financial officer David Wehner estimated the changes would cost "on the order of $10 billion" in losses during the company's Q4 2021 earnings call.

Bid subsidization increased display frequency

Beyond calculation adjustments, Meta allegedly subsidized Shops ads bids in auctions by as much as 100% to ensure more frequent display compared to non-Shops advertisements. This practice increased conversion likelihood and reinforced impressions of strong performance, according to the complaint.

The subsidization formed part of Meta's broader strategy to develop "resilient revenue" streams insulated from Apple's privacy framework. Shops ads kept tracking and transactions within Meta's applications, providing first-party data that external campaigns could not deliver.

Meta's tracking systems received additional modifications during this period. The company launched Aggregated Event Measurement (AEM1) in April 2021, using machine learning to model potential conversions while respecting users who opted out of tracking.

AEM2 rolled out shortly afterward, linking Meta app activity to third-party browsing and purchases through deterministic matching using personal identifiers including names, emails, phone numbers, addresses, and IP addresses. Purkayastha stated he believed AEM2 bypassed Apple's privacy framework restrictions, though it helped recover much of the lost tracking data.

Industry reactions and broader implications

The allegations have drawn significant attention from digital media experts and privacy advocates. Jason Kint, CEO of Digital Content Next, described the revelations as potentially representing "a BFD" (big deal) if accurate, noting the implications for both advertiser fraud complaints and privacy violations.

Kint highlighted the strategic importance of the alleged practices: "The allegations suggest facebook misled advertisers to boost Shop ad sales which mattered for two reasons: 1) revenue growth/diversification but also 2) more 1st party data as Apple privacy changes limited Facebook's tracking and 3rd party data access."

Kint also raised concerns about potential Apple policy violations, referencing the AEM2 system described in the complaint. "If these allegations from whistleblower are accurate, this is a BFD. I also don't see how Apple doesn't label this a major violation of its policies," he stated.

These allegations emerge as digital advertising faces mounting scrutiny over measurement accuracy and competitive practices. PPC Land has extensively documented concerns about inflated ROAS metrics, with industry professionals warning that campaigns showing 10x returns often perform worse than those reporting modest 2x figures due to attribution problems.

The Meta complaint also coincides with regulatory pressure on major technology platforms. Apple faced a €150 million fine from French competition authorities for implementing App Tracking Transparency in ways that harmed smaller publishers while benefiting companies with large data sets like Meta and Google.

Meanwhile, Meta has been transitioning away from on-platform commerce, ending Facebook and Instagram checkout by August 2025 and redirecting purchases to merchant websites. This shift affects how the platform calculates and reports conversion outcomes for Shops ads.

The marketing community has expressed growing concerns about platform control and transparency. Digital marketers report increased challenges with automated bidding and reduced targeting options across major advertising platforms, creating uncertainty about campaign optimization and budget allocation.

Financial impact and current status

Meta's advertising revenue reached $46.6 billion in the second quarter of 2025, representing 22% growth driven by AI-powered recommendation systems. The company's user base across Facebook, Instagram, and WhatsApp continues expanding, with Family of Apps generating 3.48 billion daily active users.

However, advertiser concerns about measurement accuracy persist. Recent research shows consumer app marketers achieve up to 214% higher ROAS when diversifying beyond Google and Meta, suggesting performance limitations within major platform ecosystems.

Purkayastha joined Meta in 2020 as a product manager on the Facebook Artificial Intelligence Applied Research team before transferring to Shops Ads in March 2022. He raised concerns about ROAS calculation methods during multiple meetings between 2022 and 2024, according to the complaint.

The filing represents part of an application for interim relief requesting reinstatement to his former position. Employment tribunal proceedings will determine whether Meta's alleged practices violated employment law and corporate transparency requirements.

Industry implications for measurement standards

The allegations highlight fundamental questions about advertising measurement consistency across platforms and products. Industry standards typically exclude shipping fees and taxes from revenue calculations to provide accurate profit margins and return metrics.

Meta has been developing value optimization tools that allow advertisers to optimize campaigns based on profit margins rather than purchase size, recognizing that certain products generate higher profitability than others. These initiatives suggest acknowledgment of measurement complexity in ecommerce advertising.

Competition authorities across multiple jurisdictions are examining whether major platforms create unfair advantages through opaque measurement practices and preferential treatment of proprietary products. The French ruling against Apple's App Tracking Transparency demonstrated regulatory willingness to challenge privacy implementations that harm competitive balance.

For marketers managing campaigns across multiple platforms, these developments underscore the importance of independent measurement and attribution systems. Third-party analytics providers continue expanding capabilities to provide unified views of cross-platform performance without relying solely on platform-reported metrics.

The outcome of Purkayastha's employment tribunal case may influence how advertising platforms disclose calculation methodologies and ensure consistent measurement standards across product lines. Industry observers will monitor whether additional whistleblower complaints emerge from other major advertising platforms facing similar measurement accuracy questions.

Timeline

• April 26, 2021: Apple releases App Tracking Transparency with iOS 14.5, reducing iOS tracking data availability
• Q4 2021: Meta CFO David Wehner estimates Apple privacy changes will cost "on the order of $10 billion" in losses
• 2020: Samujjal Purkayastha joins Meta as product manager on Facebook AI Applied Research team
• March 2022: Purkayastha transfers to Shops Ads team
• Mid-2022: Purkayastha discovers Meta counting shipping fees and taxes in Shops ads ROAS calculations
• 2022-2024: Purkayastha raises concerns about calculation methods during multiple meetings with senior leadership
• Early 2024: Meta data scientists find Shops ads ROAS inflated between 17% and 19%
• February 19, 2025: Purkayastha terminated from Meta
• March 30, 2025: French Competition Authority fines Apple €150 million for ATT implementation
• June 2025: Meta begins transitioning shops from on-platform checkout to website-based systems
• July 16, 2025: Industry experts warn about inflated ROAS metrics misleading marketers
• August 20, 2025: Purkayastha files whistleblower complaint with London Central Employment Tribunal

PPC Land explains

Return on Ad Spend (ROAS): A fundamental advertising performance metric calculated by dividing total revenue generated by advertising costs. ROAS serves as the primary indicator for campaign profitability and effectiveness across digital advertising platforms. The metric becomes problematic when attribution models incorrectly assign conversion credit or when platforms include non-merchant revenue like shipping fees in calculations, making campaigns appear more profitable than they actually perform.

Shops Ads: Meta's ecommerce advertising product designed to promote products within Facebook and Instagram's shopping ecosystem. These advertisements direct users to product pages hosted within Meta's platforms, allowing for transactions without leaving the social media environment. The product gained strategic importance following Apple's privacy changes as it provided first-party data collection opportunities that external campaigns could not deliver.

Apple's App Tracking Transparency (ATT): A privacy framework introduced with iOS 14.5 in April 2021 requiring applications to request explicit user permission before tracking activity across other companies' apps and websites. The system fundamentally disrupted digital advertising by limiting access to the Identifier for Advertisers (IDFA), forcing platforms like Meta to develop alternative tracking and attribution methods while potentially costing billions in advertising revenue.

Shipping Fees and Taxes: Additional charges applied to ecommerce transactions that do not represent revenue flowing to merchants. According to the complaint, Meta included these fees in Shops ads revenue calculations while excluding them from standard advertising products, creating an artificial performance advantage. Industry standards typically exclude such fees to provide accurate profit margins and return metrics for advertisers.

Aggregated Event Measurement (AEM): Meta's attribution system designed to track conversions while complying with Apple's privacy restrictions. AEM1 used machine learning to model potential conversions respecting user opt-out preferences, while AEM2 allegedly employed deterministic matching using personal identifiers to link activity across Meta's apps and third-party websites, potentially circumventing Apple's privacy framework.

Bid Subsidization: The alleged practice of Meta artificially increasing Shops ads bids by up to 100% in auction systems to ensure more frequent display compared to standard advertisements. This practice would increase conversion likelihood and create false impressions of superior performance while potentially violating fair competition principles in advertising auctions.

First-Party Data: Information collected directly from users through owned platforms and applications, contrasting with third-party data obtained from external sources. Following Apple's privacy changes, first-party data became increasingly valuable as platforms lost access to cross-app tracking capabilities, making products like Shops ads strategically important for maintaining targeting effectiveness.

Conversion Attribution: The process of crediting advertising touchpoints for driving desired user actions like purchases or app downloads. Attribution becomes complex in multi-touchpoint customer journeys, with different models potentially over- or under-crediting specific campaigns. Meta's alleged manipulation of attribution windows and calculation methods for Shops ads represents a significant concern for measurement accuracy.

Employment Tribunal: A UK judicial body that resolves workplace disputes between employees and employers, including cases involving whistleblower protections and wrongful termination claims. Purkayastha's filing with the London Central Employment Tribunal seeks interim relief and reinstatement while raising broader questions about corporate transparency and employee protection for those reporting alleged misconduct.

Business as Usual (BAU) Campaigns: Meta's term for traditional advertising campaigns that direct users to external websites rather than keeping them within the platform's ecosystem. These campaigns served as the primary comparison benchmark for Shops ads performance, with internal reports allegedly showing 70-80% of Shops ads outperforming BAU campaigns when inflated metrics were applied.

Five Ws Summary

Who: Samujjal Purkayastha, former Meta product manager who worked on the Shops Ads team from March 2022 to February 19, 2025, after initially joining the company in 2020 on the Facebook AI Applied Research team.

What: A whistleblower complaint alleging Meta artificially inflated return on ad spend metrics for Shops ads by 17-19% through counting shipping fees and taxes as revenue, subsidizing auction bids up to 100%, and failing to disclose calculation differences that made Shops ads appear to outperform standard advertising products.

When: The alleged practices occurred from around mid-2022 through 2024, discovered during internal reviews in early 2024, with the complaint filed on August 20, 2025, following Purkayastha's termination in February 2025.

Where: The practices allegedly occurred within Meta's Shops ads product across Facebook and Instagram platforms, with the complaint filed at the London Central Employment Tribunal as part of an application for interim relief.

Why: According to the complaint, Meta sought to boost perceived performance of its ecommerce advertising product following Apple's 2021 privacy changes that eliminated "on the order of $10 billion" in advertising revenue, requiring development of "resilient revenue" streams that kept tracking and transactions within Meta's ecosystem.