Blended ROAS vs Channel ROAS: Which Matters More

Two metrics, one question

When a client asks "how are my ads performing," they're really asking one question: is my marketing investment generating profitable returns?

Blended ROAS and channel ROAS both attempt to answer this question, but from different angles. Blended ROAS gives you the top-line answer. Channel ROAS tells you where to dig deeper.

Most agencies lean on one or the other. They'll present Meta's 4x ROAS and Google's 3x ROAS without context, or they'll show a blended 2.5x without explaining which channels are carrying the weight. Neither approach alone gives clients the information they need to make decisions.

Here's when each metric matters and how to present them together.

Blended ROAS: the truth metric

Blended ROAS = Total Revenue / Total Ad Spend

Blended ROAS is the metric that can't lie. It takes actual revenue from the client's business (Shopify, Stripe, CRM) and divides by actual total ad spend. No double-counting. No attribution inflation. No modeled conversions.

What blended ROAS tells you:

• Whether total advertising is generating a positive return.
• The trend of overall marketing efficiency over time.
• Whether scaling spend is producing proportional revenue growth.

What blended ROAS doesn't tell you:

• Which channels are contributing and which are wasting money.
• Where to shift the next dollar.
• Whether a specific campaign change improved performance.

When to use blended ROAS:

• Monthly and quarterly business reviews with the client.
• Profitability analysis and financial planning.
• Board-level reporting where the audience doesn't need channel-level detail.
• As the anchor metric that validates all other reporting.

Channel ROAS: the decision metric

Channel ROAS = Channel-Attributed Revenue / Channel Spend

Channel ROAS breaks the blended number into its component parts. It tells you how each individual channel performs relative to its cost.

But channel ROAS is only as good as the attribution model behind it. If you're using platform-reported ROAS, you're getting each platform's self-assessment. If you're using server-side multi-touch attribution, you're getting a deduplicated view.

What channel ROAS tells you:

• Which channels are performing above or below average.
• Where to shift budget for marginal improvement.
• Which campaigns or ad sets within a channel are strongest.

What channel ROAS doesn't tell you:

• Whether the attributed revenue is truly incremental.
• How channels interact to produce conversions.
• Whether you're comparing apples to apples (attribution methods differ by channel).

When to use channel ROAS:

• Weekly performance reviews with the media buying team.
• Budget reallocation discussions between channels.
• Campaign-level optimization within each channel.

The gap between them tells a story

The relationship between blended ROAS and the sum of channel ROAS reveals important dynamics:

If the weighted average of channel ROAS is much higher than blended ROAS: Significant double-counting is occurring. Multiple channels are claiming the same conversions. This is normal when using platform-reported metrics, but the gap size tells you how much inflation exists.

If the weighted average of channel ROAS is close to blended ROAS: Your attribution model is doing a reasonable job of deduplicating. Or you have minimal channel overlap (possible with distinct audience targeting).

If one channel's ROAS is significantly higher than all others: That channel is either genuinely outperforming or it's benefiting from favorable attribution (e.g., branded search captures demand created by other channels). Investigate with incrementality testing before scaling.

If blended ROAS is declining but all channel ROAS metrics are stable: You have a measurement problem, not a performance problem. New spend isn't being captured by any platform's attribution. Or increasing channel overlap is inflating channel metrics while the blended number reveals the truth.

How agencies should present both metrics

The executive summary: lead with blended

Start every client report with blended ROAS. It's the number that ties to their P&L and can't be disputed.

"This month, total ad spend was $85,000 and total revenue was $212,000. Blended ROAS is 2.5x, up from 2.3x last month. Your advertising is profitable."

This anchors the conversation in reality before diving into channel details.

The channel breakdown: show attributed, not platform

Present channel ROAS using your attribution model, not each platform's self-reported numbers. Show the numbers in a single table with consistent methodology.

| Channel | Spend | Attributed Revenue | Channel ROAS | |---------|-------|--------------------|-------------| | Meta Prospecting | $35,000 | $70,000 | 2.0x | | Meta Retargeting | $10,000 | $32,000 | 3.2x | | Google Non-Brand | $20,000 | $52,000 | 2.6x | | Google Branded | $8,000 | $28,000 | 3.5x | | TikTok | $12,000 | $30,000 | 2.5x | | Total | $85,000 | $212,000 | 2.5x |

Notice that the attributed revenue sums to total revenue (because the attribution model deduplicates). This is the power of a unified approach.

The insight layer: explain the dynamics

Don't just show numbers -- explain what they mean and what to do about them.

"Google Branded has the highest channel ROAS at 3.5x, but this channel primarily captures demand created by other channels. Scaling branded search spend won't proportionally increase revenue. The highest marginal return is in Meta Prospecting and TikTok, where we're still in the efficient part of the response curve."

The recommendation: grounded in both metrics

"I recommend shifting $5,000 from Meta Retargeting to TikTok. Retargeting is showing high ROAS but incrementality testing last quarter showed it has significant overlap with email and organic conversions. TikTok is generating genuinely new customer demand with a healthy 2.5x ROAS. This shift should maintain or improve blended ROAS while expanding new customer acquisition."

Common arguments and honest answers

"But Meta says our ROAS is 4x, not 2x." Meta's ROAS counts all conversions that happen within its attribution window, including view-throughs and conversions that other channels also claim. The 2x figure uses deduplicated attribution from server-side data. Both are mathematically correct -- they just measure different things. We use the deduplicated number for budget decisions because it aligns with your actual revenue.

"Why is our blended ROAS lower than every channel's ROAS?" Because platform-reported channel ROAS double-counts shared conversions. If 30% of conversions are claimed by both Meta and Google, each channel's ROAS is inflated while the blended number reflects the single-counted reality.

"Should we cut the channel with the lowest ROAS?" Not necessarily. The channel with the lowest ROAS might be generating the highest volume of new customers at an acceptable cost. Or it might be an awareness channel that enables other channels to convert. Evaluate channel value based on incremental impact, not just ROAS.

The hierarchy of ROAS metrics

For agency decision-making, use this hierarchy:

1. Blended ROAS for overall health and trend monitoring.
2. Attributed channel ROAS (from unified attribution) for budget allocation between channels.
3. Incremental ROAS (from experiments) for validating whether a channel truly earns its budget.
4. Platform ROAS for within-channel campaign and creative optimization.

Each level answers different questions at different levels of rigor. Using the wrong metric for the wrong decision is where agencies go wrong.

FAQ

What's a good blended ROAS target for agencies to set with clients?

It depends entirely on the client's gross margins and repeat purchase economics. A 2x blended ROAS is excellent for a brand with 70% gross margins and 40% 12-month repeat rates. It's unsustainable for a brand with 40% margins and low repeat rates. Work backward from profitability: Minimum ROAS = 1 / (Gross Margin x (1 + Expected Repeat Revenue Factor)). Set the target above this floor.

How often should I report blended vs. channel ROAS?

Report blended ROAS weekly as a quick health check and monthly in detail. Report channel ROAS monthly with attribution-adjusted numbers. Report incremental ROAS quarterly based on testing cycles. Weekly channel ROAS is useful for your internal team but too granular for most client reports.

What if a client only cares about channel ROAS and won't look at blended?

Show them the math. Pull their actual revenue from Shopify, divide by total spend, and present it alongside the sum of platform-reported revenue. When they see that platform reports claim 2.5x their actual revenue, they'll understand why blended ROAS exists. If they still insist on platform numbers, document your recommendation in writing and report both.

---

Go Funnel uses server-side tracking and multi-touch attribution to show you which ads actually drive revenue. Book a call to see your real numbers.