Why Your ROAS is Lying to You (And How to Fix It)

Your Meta Ads dashboard says 4.2x ROAS. Your Google Ads says 3.8x. Revenue looks good. But when you look at your actual bank balance, the numbers do not add up. You are spending more on ads than ever but profit has not moved. Sound familiar? Your ROAS is lying to you — and the fix is simpler than most advertisers realise.

The Attribution Problem in Plain Language

ROAS as reported by advertising platforms is almost always inflated — sometimes significantly. The reason is attribution: every platform counts the conversions it can claim credit for, and multiple platforms routinely claim credit for the same conversion.

Here is a scenario we see constantly. A customer sees your Meta ad on Monday and scrolls past it. They click a Google search ad on Wednesday after searching for your brand. They read a blog post on Thursday via organic search. On Friday they go directly to your website and purchase. Meta counts this as a conversion via view-through attribution. Google counts it as a conversion via last-click. If you sent an email newsletter that week, your email platform might count it too. One sale, three platforms each reporting a conversion, each inflating your apparent ROAS.

Why Platform ROAS Will Always Be Inflated

Platform-reported ROAS is not a neutral metric. Each platform's attribution model is designed, at least in part, to justify its own value. Meta's default attribution window is a 7-day click and 1-day view. This means Meta counts a conversion if someone clicked your ad in the past seven days OR merely viewed it in the past 24 hours — regardless of what else happened in between. Google's last-click attribution gives 100% credit to the final click before purchase, ignoring all other touchpoints.

Neither model accurately represents how customers actually buy. Most purchases involve multiple touchpoints across multiple days and multiple channels. No single attribution model can accurately represent this complexity, which is why any individual platform's ROAS number will always be directionally useful but numerically unreliable.

The Real Metric: Blended ROAS (MER)

The most reliable performance metric in e-commerce and direct response advertising is blended ROAS, also called Marketing Efficiency Ratio (MER). The calculation is simple: total revenue from all sources divided by total ad spend across all paid channels. No attribution windows. No view-through conversions. No platform-specific models. Real money in, divided by real money out.

If you are spending $15,000 per month across Meta, Google, and email and generating $75,000 in revenue, your blended ROAS is 5x — regardless of what any individual platform reports. This is your north star metric. Track it weekly. Every significant change in blended ROAS tells you something real is happening in your business.

How to Build a Reliable Attribution System

Step 1: Implement Blended ROAS tracking immediately. Connect your ad platform spend data and actual revenue data in a single spreadsheet or dashboard. Calculate total spend and total revenue weekly. This takes 30 minutes to set up and becomes your most reliable signal.

Step 2: Install Conversions API for Meta. Browser-based Meta Pixel alone misses 20-40% of conversions in 2026 due to iOS privacy restrictions and browser-level tracking prevention. Conversions API sends server-side conversion data directly to Meta, recovering a significant portion of lost attribution and dramatically improving campaign optimisation.

Step 3: Use UTM parameters on every ad. Every ad across every platform should have consistent UTM parameters: utm_source, utm_medium, utm_campaign, utm_content. This allows Google Analytics 4 to independently attribute conversions by traffic source, giving you a cross-channel view that is not controlled by any individual platform.

Step 4: Switch to data-driven attribution in GA4. Google Analytics 4's default attribution model is last click, which undervalues top-of-funnel channels like display and social. Data-driven attribution uses machine learning to distribute credit across all touchpoints based on observed conversion patterns. It is not perfect but it is significantly more accurate than last click.

Step 5: Run incrementality tests quarterly. The most accurate way to understand a channel's real contribution is to pause it and measure the impact on total revenue. Pause Meta for two weeks and measure whether blended ROAS improves, stays flat, or declines. Pause Google and repeat. This reveals the true incremental contribution of each channel — something no attribution model can fully capture.

How to Interpret the Results

Most businesses we work with find that their actual blended ROAS is 30-50% lower than the average of what their individual platforms report. This is not a sign that the advertising is not working — it is a sign that the attribution was overcounting. The practical implication is that budget decisions based on platform-reported ROAS tend to over-invest in channels that are good at claiming credit (often last-click channels like branded search) and under-invest in channels that influence decisions earlier in the funnel (often Meta and display).