The Deceptive Allure of Inflated ROAS: How Misconfigured Conversions Distort Performance Metrics and Misguide Marketing Spend

A recent analysis of digital advertising performance has unearthed a pervasive issue: the widespread misrepresentation of Return on Ad Spend (ROAS) due to flawed conversion tracking and valuation within Google Ads accounts. This phenomenon, while often stemming from unintentional configuration errors, can lead to wildly inaccurate performance metrics, potentially misdirecting significant marketing budgets and creating a false sense of success for businesses. The implications are far-reaching, impacting campaign optimization, strategic decision-making, and ultimately, profitability.

The core of the problem lies in how Google Ads attributes value to conversions. While the platform is designed to optimize campaigns based on reported performance, a fundamental misunderstanding or misconfiguration of conversion settings can lead to the system being fed incorrect data. This was starkly illustrated in an anecdote shared by an industry expert, where a student presented a Google Ads report showcasing an astonishing ROAS exceeding 500%, yet simultaneously reported zero actual sales. This apparent paradox immediately flagged a critical issue: Google Ads was likely counting something as a sale that did not represent a genuine transaction or valuable lead.

"If your ROAS is big and beautiful, Google thinks you have sales. If you have no sales, Google is counting something as a sale that isn’t one," the expert explained, pinpointing conversion settings as the immediate area for investigation. While this particular case was an extreme example of a "total lie," the expert warns that more insidious problems exist in accounts where ads do generate some sales, but the reported numbers are only partially accurate. In such scenarios, a flawed setting can masquerade as success, creating a misleading impression of a "good year" for the business.

This issue is a critical, often overlooked, layer beneath more commonly discussed problems like attribution models or geo-holdout testing. Before delving into complex analytical methods, advertisers are urged to first verify if their reported numbers are even being measured accurately. The findings suggest that in a significant portion of analyzed accounts, a critical part of the conversion value is not derived from actual transactions or verified leads, but rather from manually entered figures.

The Human Element in Digital Valuation: Static Values and Their Pitfalls

A significant contributor to inflated ROAS figures is the use of static values for conversion actions. Every conversion action configured with a fixed monetary value, rather than dynamically pulling revenue from actual sales, relies on a number manually entered by an individual. Google Ads then aggregates these static values with genuine revenue from completed orders, creating a combined "conversion value" that is then divided by advertising cost to calculate ROAS. The issue is that the reporting interface does not clearly distinguish between measured revenue and these manually imputed figures.

Consider a hypothetical scenario: an e-commerce business generates 1,000 orders worth $100,000 and 2,000 leads each quarter, with an advertising spend of $20,000. If each lead is assigned a static value of $10, the reported conversion value would be $120,000, yielding a ROAS of 6.0 ($120,000 / $20,000). Now, imagine that before the next quarter begins, a marketing manager decides that leads are actually more valuable and updates the static value to $20. Crucially, this change only applies to new conversions generated from that point forward.

The subsequent quarter might yield precisely the same number of orders, the same quantity of leads, and the same revenue from sales. However, the reported conversion value would jump to $140,000 (original $100,000 from sales + 2,000 leads * $20 value), resulting in a reported ROAS of 7.0 ($140,000 / $20,000). In this instance, a seemingly positive ROAS increase of one-sixth was not driven by improved sales performance or increased lead quality, but solely by a person altering a single number in a settings field.

This dynamic is particularly problematic when comparing performance across different periods. If the static value for leads changes midway through a comparative analysis period (e.g., comparing Q1 this year to Q1 last year), the ROAS figures are being measured in fundamentally different units. While a chart might present these disparate metrics as a continuous line, it creates a misleading representation of growth and efficiency. The integrity of historical performance data is compromised, making accurate year-over-year comparisons unreliable.

The Multiplication of Success: Overcounting Leads and Duplicate Conversions

Beyond the issue of static values, the way conversions are counted can also artificially inflate performance metrics. A common error identified in approximately four out of every five accounts analyzed that utilize leads for bidding is the "Every" conversion setting applied to lead-related actions. Instead of counting one unique lead per ad click, this setting counts every instance a conversion action is triggered.

For actual purchases, the "Every" setting can be appropriate, as a single customer might indeed make multiple purchases. However, for lead generation, Google’s own guidance suggests that typically, only one unique lead per ad click adds value. When the "Every" setting is applied to a lead form, a user who submits the form, notices a typo in their phone number, and resubmits it, will be counted as two separate leads. Similarly, if a thank-you page loads twice due to a technical glitch, it can result in two counted leads from a single user interaction. This effectively doubles or even triples the perceived number of leads generated from a single marketing touchpoint.

A second, and equally concerning, source of inflated conversion numbers stems from the existence of duplicate conversion counters. More than one in four accounts with purchase tracking have two or more distinct purchase actions within the "conversions" column. While sometimes these may track genuinely different sales channels or transaction types, it is often the case that they are counting the same order multiple times. This duplication can occur through various means, such as tracking an order via the Google tag on the website and simultaneously importing the same order data through a Google Analytics 4 (GA4) import or a Customer Relationship Management (CRM) upload.

The danger here is not necessarily in the presence of a faulty counter that drifts significantly from reality – these are often recognized as broken and removed. The truly perilous duplicates are those that closely mirror the backend system’s accurate sales data. Because they appear to be functioning correctly, they are often left untouched. These seemingly healthy, yet redundant, counters can persist as secondary actions for months. The moment such a duplicate is added to a campaign’s conversion goals, the reported conversion value can effectively double overnight, impacting the bidding strategy of every campaign reliant on that goal. Alarmingly, no alerts are triggered because, from Google Ads’ perspective, nothing has technically "broken."

Unused Potential and Misdirected Optimization: Primary Actions and Phantom Goals

A significant portion of analyzed accounts, approximately two out of every three, are found to retain at least one conversion action marked as "primary" that is not actually being utilized by any active campaign goals. These actions carry the designation of importance within the account settings but are invisible to campaigns that are actively bidding. Consequently, they are often overlooked during routine account management.

The origins of these unused primary actions are frequently lost to time. They might have been set up for a specific, now-defunct campaign, or created as a custom goal in haste during a previous optimization effort. The critical issue arises when a campaign goal is later modified, or a new custom goal is constructed, and this dormant primary action is inadvertently included. Smart Bidding algorithms, designed to maximize campaign performance based on the defined goals, will then begin to optimize for an event that has not been actively monitored or validated for years. Google’s own documentation explicitly states that secondary actions included in custom goals are utilized for bidding once that goal is set within a campaign. Audits rarely reveal a systematic check of these custom goals, leaving this potential for misdirection unaddressed.

The Siren Song of Cheap Clicks: Soft Events and Their Costly Influence

Perhaps one of the most insidious ways performance metrics can be distorted is through the inclusion of "soft" conversion events within the primary conversion column that campaigns bid on. In roughly six out of every ten accounts reviewed, actions such as page views, engagement metrics, file downloads, outbound clicks, direction requests, or follow-on video views are present in the conversions column alongside genuine sales or qualified leads.

These "soft" events, unlike genuine sales or qualified leads, are often triggered with far greater ease and speed. Cheap placements, such as display ads or low-intent search queries, are far more efficient at generating these types of interactions than a search ad is at producing a high-value order. Consequently, Smart Bidding algorithms, seeking to maximize conversions within their defined parameters, will incrementally shift budget towards these placements and actions that are most adept at "filling the counter." The reported outcome is often framed as improved efficiency, but it is an efficiency built on a foundation of less valuable, or even valueless, interactions.

There is a rare, honest exception to this rule. For very small lead generation accounts where genuine leads are too infrequent for bidding algorithms to learn effectively, a softer event might be deliberately used as a proxy. However, this approach is only effective when it is a conscious, strategic decision, with the proxy event assigned a value that accurately reflects its true conversion rate to a sale multiplied by the average sale value. In the vast majority of cases, these soft events found within the primary conversion column were never the result of a deliberate strategic choice.

This problematic inclusion of soft events is also a primary reason why setting a Target ROAS on an account with unreliable conversion counting is ill-advised. When the underlying data is flawed, a ROAS target does not protect a business’s profit margins; instead, it provides the algorithm with a precise instruction based on inaccurate numbers, potentially leading to significant financial losses.

The Preemptive Strike: Verifying Measurement Before Testing

In light of these findings, industry professionals advocate for a foundational step before engaging in more complex performance analysis or testing. The recent publication of a comprehensive guide to incrementality testing by PPC Hero, which rightly emphasizes measuring lift in actual store sales rather than solely relying on Google Ads metrics, highlights this need. However, the crucial point is that the bidding and budget discussions surrounding such tests are still heavily influenced by the data presented in the conversions column.

Therefore, the expert recommendation is clear: fix the conversion column first. This is a relatively low-cost endeavor, often requiring little more than an afternoon’s dedicated effort. The proposed methodology involves a systematic review and clean-up of conversion settings.

1. Eliminate duplicate conversion actions: Scrutinize the "conversions" column for any identical or near-identical actions. Remove any duplicates, prioritizing the most accurate and reliably tracked source.
2. Review the "Count" setting for lead actions: For any lead-related conversion actions, ensure the "Count" setting is appropriately configured to "One" rather than "Every," unless there is a specific, justifiable reason for counting multiple instances per click.
3. Audit static conversion values: For conversion actions with static values, verify that these values accurately reflect the true lead-to-sale rate multiplied by the average sale value. Critically, ensure these values have not been arbitrarily changed within the reporting period being analyzed. Consider replacing static values with dynamic revenue tracking from actual sales whenever possible.
4. Identify and remove "soft" conversion events: Remove any non-transactional events like page views, file downloads, or outbound clicks from the primary conversion column unless they have been deliberately implemented as a strategic proxy with an accurately assigned value.
5. Assess primary conversion actions for campaign usage: Review all conversion actions marked as "primary" and confirm that they are actively included in campaign goals. Deactivate or reconfigure any primary actions that are not being utilized for bidding.

What survives this rigorous clean-up process is a set of metrics that represent genuine business outcomes. These are the numbers that should be defended in budget meetings and serve as the basis for incrementality testing. Growth figures that can be reliably compared year over year, based on accurately measured revenue and valuable conversions, are the true indicators of marketing success.

The Bottom Line: Accuracy is the Bedrock of Performance

Ultimately, a ROAS figure is only a meaningful measurement when every component contributing to it has been accurately measured. A static value entered by a person is an estimate from a specific point in time, subject to change and potential bias. A second, duplicate counter adds no new information but artificially inflates the perceived volume of success. Both of these misconfigurations can become exceedingly expensive when they form the basis of campaign optimization and budget allocation decisions.

The student who reported a 500%+ ROAS with no sales had an error so blatant it was impossible to miss. However, in most accounts, the misrepresentations are subtler, presenting numbers that are not only believable but desirable. This creates a dangerous illusion of success that can lead to sustained underperformance and wasted advertising expenditure.

To foster transparency and accountability, a crucial recommendation for the upcoming months is to report order revenue and any typed-in lead values as separate line items. This simple step will empower stakeholders to clearly distinguish between actual, measured revenue and estimated values, ensuring that ROAS calculations are based on concrete data rather than potentially misleading assumptions. Only through rigorous attention to the foundational elements of conversion tracking can businesses ensure their digital advertising efforts are truly driving profitable growth.