MER vs ROAS: Which Metric Should You Optimize

Two metrics, different philosophies

The debate between MER and ROAS reflects a deeper disagreement in marketing measurement: should you optimize the parts or optimize the whole?

ROAS says optimize the parts. Measure each channel's return, scale the winners, cut the losers. It's precise, actionable, and familiar to every media buyer.

MER says optimize the whole. Measure total revenue against total spend. Don't worry about which channel gets credit -- just grow the overall ratio. It's simple, un-gameable, and increasingly popular.

Both have legitimate uses. Neither is sufficient alone. Here's how they work, when each matters, and how to use them in combination.

Marketing Efficiency Ratio (MER) explained

MER = Total Revenue / Total Marketing Spend

MER doesn't use attribution at all. It takes total business revenue (from Shopify, Stripe, or your accounting system) and divides by total marketing spend (every dollar spent on acquiring and retaining customers).

Example:

• Monthly revenue: $500,000
• Total marketing spend: $120,000 (includes Meta, Google, TikTok, email platform, influencer payments, agency fees)
• MER: 4.17x

What MER captures:

• The holistic efficiency of your entire marketing machine.
• Cross-channel effects (a Meta ad that drives a Google search that drives an email conversion -- all captured).
• Organic and brand effects (strong brand equity reduces the need for paid spend, improving MER).
• Attribution-proof performance (no platform can inflate MER because it uses actual revenue).

What MER misses:

• Channel-level performance. A 4x MER doesn't tell you whether Meta is pulling its weight or whether Google is carrying the team.
• Actionability for media buyers. If MER drops from 4x to 3.5x, what do you change? MER can't point you to the problem.
• Causal relationships. Revenue could be declining for reasons unrelated to marketing (product issues, seasonal dip, economy). MER attributes all revenue changes to marketing efficiency.

Return on Ad Spend (ROAS) explained

ROAS = Channel-Attributed Revenue / Channel Ad Spend

ROAS measures the return on a specific channel's ad investment. It requires an attribution model to assign revenue to channels.

Three flavors of ROAS:

1. Platform ROAS: Revenue as reported by Meta, Google, etc. Inflated by double-counting and view-through attribution. Useful for within-channel optimization.

2. Attributed ROAS: Revenue assigned by a unified attribution model (server-side, multi-touch). Deduplicated across channels. More accurate for cross-channel comparison.

3. Incremental ROAS: Revenue that would not have occurred without the channel, measured via experiments. The most rigorous definition. Requires testing to calculate.

What ROAS captures:

• Channel-level performance differences.
• Campaign and creative-level signals for optimization.
• Clear targets for media buyers to manage against.

What ROAS misses (depending on the flavor):

• Platform ROAS: massively over-credits due to double-counting.
• Attributed ROAS: depends on attribution model quality; can still misallocate credit.
• Incremental ROAS: accurate but expensive and slow to produce (requires experiments).

When MER is the right metric

Scenario 1: Attribution is broken or unreliable

If your tracking has gaps, your attribution model is questionable, or you've recently changed measurement methodology, channel ROAS is unreliable. MER provides a stable, trustworthy signal while you fix attribution.

Scenario 2: You're evaluating total marketing investment

When the CEO asks "Should we spend $100K or $150K on marketing next month?", MER answers the question directly. "At current MER of 4x, $150K in spend should produce approximately $600K in revenue." ROAS can't answer this because it doesn't address the total picture.

Scenario 3: You're running many channels with heavy overlap

If you run Meta, Google, TikTok, CTV, email, influencer, and SEO -- with significant audience overlap -- channel ROAS is inherently unreliable (too much double-counting). MER bypasses the attribution problem entirely.

Scenario 4: Brand-heavy marketing

If a significant portion of your marketing is brand-building (TV, podcasts, sponsorships, PR), these channels often don't appear in attribution at all. MER captures their impact because it measures total revenue, not just attributed revenue.

When ROAS is the right metric

Scenario 1: Day-to-day media buying

Media buyers need channel and campaign-level signals to make decisions. "Pause this ad set." "Scale this campaign." "Shift budget from Display to Search." These decisions require channel ROAS, not blended MER.

Scenario 2: Budget allocation between channels

If you have $100K to allocate across 5 channels, you need to know which channels have the highest marginal return. Channel ROAS (ideally attributed or incremental) provides the directional signal for allocation.

Scenario 3: New channel evaluation

When testing a new channel (launching TikTok, adding CTV), you need channel-specific performance data to decide whether to continue investing. MER can't isolate the new channel's impact from the rest of the mix.

Scenario 4: Performance troubleshooting

If blended MER drops, you need channel ROAS to diagnose the cause. Is Meta declining? Is Google oversaturated? Is TikTok underperforming? Without channel-level data, you can't identify or fix the problem.

Using MER and ROAS together

The most effective approach uses both metrics at different levels of the decision hierarchy:

Level 1: MER as the health check

Track MER weekly. It's your vital sign. If MER is stable or improving, the marketing machine is healthy regardless of what individual channel reports say. If MER is declining, something needs attention.

MER targets by business stage:

• Early-stage (acquiring first customers): 2-3x is acceptable. You're investing heavily in acquisition.
• Growth-stage (scaling profitably): 3-5x is the target range.
• Mature (optimizing efficiency): 5-8x indicates a well-tuned machine.

Level 2: Attributed ROAS for allocation

Use attributed ROAS (from server-side multi-touch attribution) to distribute budget across channels. This is the metric your media buying team optimizes against for weekly and monthly allocation decisions.

How it connects to MER: If you reallocate budget based on attributed ROAS and MER improves, the reallocation was correct. If MER declines despite favorable attributed ROAS changes, your attribution model may be misleading you.

Level 3: Incremental ROAS for validation

Run incrementality tests quarterly to validate that your attributed ROAS estimates are directionally correct. If attributed ROAS says Meta is 2.5x and an experiment says 2.2x, the attribution is well-calibrated. If the attributed ROAS says 2.5x and the experiment says 0.8x, your attribution model needs adjustment.

Level 4: Platform ROAS for channel optimization

Use platform-reported ROAS within each channel for campaign-level decisions. Which campaigns to scale, which creatives to test, which audiences to expand. Within-channel relative comparisons are valid even when absolute numbers are inflated.

The practical workflow

Here's how a media buyer should use both metrics in a typical week:

Monday: Check MER for the previous week. Is it in line with the target? Trending up or down?

Tuesday-Thursday: Optimize campaigns using platform ROAS. Pause low performers, scale high performers, test new creatives.

Friday: Review attributed ROAS by channel. Any channels significantly above or below target? Flag for potential reallocation.

Monthly: Reconcile attributed ROAS with MER. If there's a growing gap, investigate. Run incrementality tests on the most uncertain channels.

Quarterly: Review incremental ROAS results. Calibrate the attribution model. Adjust MER targets based on the latest data.

FAQ

Can MER be misleading?

Yes. MER can improve for non-marketing reasons -- a viral moment, seasonal tailwind, or product launch. If revenue spikes without any marketing change, MER improves, and you might wrongly conclude your marketing is more efficient. Always triangulate MER with attributed ROAS and common sense. If MER improved but you didn't change your marketing, look for external explanations.

What if my MER is great but my channel ROAS is low?

This usually means a significant portion of revenue is coming from organic and brand sources (direct visits, organic search, word of mouth) that aren't attributed to any paid channel. Your paid channels might be efficient at creating awareness that converts through organic paths. Or you might be spending on channels that aren't contributing -- MER alone can't tell you which. Use incrementality testing to resolve the ambiguity.

Should agencies report MER or ROAS to clients?

Both. Lead with MER as the headline metric because it ties to the client's business results and can't be disputed. Follow with attributed channel ROAS for actionable channel-level insights. If you only report channel ROAS, the client will eventually notice the numbers don't add up to their actual revenue. If you only report MER, the client won't know what you're doing with their budget at the channel level. Both metrics, together, build trust.

---

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.