Why Your ROAS Number Lies (2026 Edition)
Written by Elias Oender
May 7, 2026 3 min read
The quick answer
ROAS (Return on Ad Spend) is increasingly unreliable in 2026 due to AI-driven inflation of platform-reported conversions. Contribution margin, which accounts for product costs, operational expenses, and customer lifetime value, provides a truer measure of profitability. To navigate this, marketers should focus on attribution models that capture full-funnel impact and prioritize contribution margin over raw ROAS.
Why Is ROAS More Misleading Than Ever in 2026?
In 2026, ROAS (Return on Ad Spend) is a vanity metric. Platforms like Google and Meta, armed with AI-powered tools like Gemini and Meta’s AI connectors, are optimizing for clicks and engagements, not necessarily profitable sales. These tools inflate reported conversions, making your ROAS look stellar while masking underlying inefficiencies, as covered here.
For example, Google’s new Business Agent for Leads and Meta’s AI business assistant can drive more leads, but how many of those leads are qualified? How many convert into paying customers? The numbers might look good, but the truth is often less impressive.
What’s Wrong With ROAS Alone?
ROAS doesn’t account for product costs, shipping, discounts, or operational expenses. It’s a top-line metric that ignores the bottom line. A high ROAS might mean you’re generating revenue, but not necessarily profit. For instance, if your ROAS is 5x, but your contribution margin is negative, you’re losing money with every sale, as explained here.
Contribution margin, revenue minus variable costs, tells the real story. It’s the metric that shows whether your ads are actually driving profitable growth.
How AI Inflates Platform-Reported Conversions
The shift from “AI that answers” to “AI that gets things done” has created a new challenge. Platforms are incentivized to report conversions that align with their own goals, not yours. For example:
- Google’s Smart Bidding Exploration now extends to Performance Max and Shopping campaigns, optimizing for clicks that may not lead to sales.
- Meta’s AI connectors allow advertisers to run ads through external workflows, but these workflows can prioritize engagement over profitability.
As a result, platform-reported conversions often don’t align with real-world outcomes. This disconnect makes ROAS an unreliable benchmark, as highlighted in this analysis.
One account we manage solves this by logging conversions as hard (booked calls, verified leads) vs soft (forms, newsletters), with cross-session contacts tracked manually. This ensures no single event swings a decision and ROAS reflects real outcomes, not platform-reported noise. For example, a high ROAS campaign was scaled down after manual tracking revealed most leads were unqualified, despite platform metrics claiming success.
What Should You Measure Instead?
Focus on contribution margin and attribution models that capture full-funnel impact. Here’s how:
- Audit Your Conversions: Compare platform-reported conversions to your own CRM data. Look for discrepancies and adjust your tracking.
- Calculate Contribution Margin: Include all variable costs in your profitability analysis. This gives you a clearer picture of ad performance.
- Use Multi-Touch Attribution: Understand how ads contribute to the entire customer journey, not just the last click.
For more on optimizing your campaigns, run a free scan to identify gaps in your strategy.
The Bottom Line
In 2026, ROAS is a trap. Platforms are gaming the numbers, and AI is making it easier than ever to inflate reported performance. Contribution margin is the metric that matters. By focusing on profitability, not just revenue, you can make smarter decisions and drive real growth.
Need help navigating these changes? Book a call to discuss how to build a strategy that works for your business.
