MER is Marketing Efficiency Ratio: revenue divided by the marketing spend included in your definition. It gives a business level comparison alongside the platform reports. The useful work is keeping the definitions consistent and checking the result against margin.
The formula
Blended MER equals total revenue divided by total paid marketing spend. The formula uses the recorded totals for a defined period. Revenue is the recorded Shopify revenue under the order and refund rules you selected. Spend is every dollar you paid to Meta, TikTok, Google, LinkedIn, influencer deals, sponsorships, and anything else with an invoice that was meant to drive revenue.
$300K in Shopify revenue divided by $160K in paid marketing spend is a blended MER of 1.875x. The fact that Meta claims 4.1x ROAS on its share, TikTok claims 2.6x on its share, and Google claims 3.8x on its share is a separate conversation from what the business actually did. See the budget example below for how overlapping platform claims answer a different question.
What to include in spend
This is where brands argue themselves out of using MER. The temptation is to exclude certain channels because "they're not really the same as paid social." Resist.
Include in paid marketing spend:
- Platform ad spend (Meta, TikTok, Google, LinkedIn, Pinterest, Reddit, anywhere you buy media)
- Influencer deals paid in cash
- Affiliate payouts
- Sponsorships and brand deals
- Creator gifting at book value
- Email and SMS platform fees (Klaviyo, Attentive, etc.)
- Agency retainers and freelancer media fees (debated but usually include)
Do not include:
- Tool subscriptions that are not channel-specific (Slack, Notion, etc.)
- Salary of internal marketing staff
- Product cost or COGS
The debate usually centers on agency fees. My call: include them. The agency fee is marketing spend directed at producing revenue. Excluding it flatters the MER number and hides the total cost of your marketing operation.
What to include in revenue
Top-line Shopify revenue is the simplest baseline. Some operators argue for net revenue (after returns), gross profit, or contribution margin. All are defensible. My call for most brands: use gross revenue as the top-line MER number and track contribution margin MER separately as a cross-check.
The MER threshold by gross margin
The minimum viable MER depends on your gross margin. Here is the math.
In a simplified model with gross margin as the only allowance before marketing, the break even MER is 1 divided by the gross margin fraction. That gives floors of 2.0 at 50 percent gross margin, 2.86 at 35 percent, and 4.0 at 25 percent.
These examples cover the marketing spend after cost of goods in that simplified model. Fulfillment, transaction fees, fixed overhead, and the required profit buffer can raise the target. A brand with 40 percent gross margin running MER of 2.0x looks healthy in isolation but is actually losing money on contribution. The floor for that brand is 2.5x. Any MER under 2.5x is unprofitable.
This math is what agencies rarely foreground. Platform ROAS of 4x sounds great until you do the gross margin adjustment and realize the blended math is underwater. Why agencies overspend on top of funnel covers the structural reason this math gets buried.
A budget change that can fool a platform report
Consider a hypothetical account where Meta reports 4x ROAS and TikTok reports 2x. Moving spend from TikTok to Meta looks reasonable in those reports. If TikTok was introducing customers who later bought through branded search or retargeting, cutting it could reduce total sales while Meta continues claiming credit for the customers who remain.
That outcome is possible, rather than a rule about either platform. TikTok could also be the weaker use of the budget. Keep the example as a reason to test the contribution instead of assuming that the lower platform number identifies the channel to cut.
Overlapping attribution means the same order can appear in several channel reports. Summing attributed revenue can therefore exceed the order ledger. The size and direction of that difference depend on the reporting setup; the older article's 30 to 60 percent range was an observation, not a benchmark for every store.
What MER can establish
MER connects the spend definition to the revenue definition. It still cannot assign causal credit to one channel. Seasonality, price changes, returning customers, stock availability, and promotions can move the ratio while media quality stays unchanged.
A channel pause is an intervention, but a simple before and after comparison can confuse those effects with the pause itself. Where practical, use a randomized holdout or a carefully designed geographic test with a comparison group, a defined measurement period, and uncertainty reported alongside the estimate. Agree on the downside and stopping rule before reducing spend.
The incrementality test guide explains how to investigate channel contribution. Platform ROAS remains useful for locating a reporting change; the business ratio and a suitable test answer the broader budget questions.
The monthly reporting rhythm
Every month, produce one page. Top of the page: three numbers.
- Total revenue
- Total paid marketing spend (all channels, invoiced this month)
- Blended MER (revenue divided by spend)
Underneath sits the comparison: how does this month's MER compare to the last three months, the last six months, and the same month last year? Trend matters more than the spot number.
Then comes the platform diagnostic section. Meta ROAS, TikTok ROAS, Google ROAS. Each labeled clearly as "platform-attributed, diagnostic." Not as the north star. These numbers are there so you can identify which platform is drifting when blended MER moves.
Include the results and limits of any incrementality test run during the period. A channel report and a business efficiency ratio should be read alongside that causal evidence.
Weekly vs. monthly cadence
Run MER monthly as the primary rhythm. Weekly MER is too noisy for most DTC brands. Events like a single affiliate post, a single viral TikTok, or a single Meta algorithm hiccup can swing weekly MER by 30 percent without representing a real business change. Monthly smooths the noise.
If you are running at $500K+/month paid spend, biweekly MER starts making sense because the sample size is larger. Still not weekly.
Breaking down MER by campaign stage
Blended MER is the whole-business number. It hides important detail. A useful secondary cut: MER by campaign stage.
- TOF MER: revenue attributed to top-of-funnel cohorts divided by TOF spend
- MOF MER: revenue attributed to middle-of-funnel audiences divided by MOF spend
- Retargeting MER: revenue from retargeting cohorts divided by retargeting spend
These cuts require a warehouse or at least honest attribution modeling in a BI tool. Platform ROAS per campaign stage is not the same thing because it still relies on platform attribution claims. The cleanest version uses a warehouse-based attribution model that reconciles first-party session data with Shopify orders.
“MER is the only paid social number that survives the translation to a board meeting. Platform ROAS does not. If your metric cannot be explained to a non-marketer in 30 seconds, it is the wrong metric.
”
The CAPI connection
Blended MER does not care about CAPI quality in its top-line formula. Your revenue number is your revenue number. But the platform-level diagnostics you use to investigate MER movements depend entirely on CAPI quality. If Meta's reported ROAS is moving because match quality is dropping, you cannot tell the difference between a genuine performance problem and a measurement problem without clean CAPI. Wiring CAPI signal back into the creative testing loop covers the CAPI side of this work.
The trap operators fall into
The trap is this: once you start measuring on MER, the platform ROAS numbers look bad, and the temptation is to explain them away. "Our Meta ROAS is down but that's because iOS 17 hit harder this quarter." "TikTok ROAS looks weak but the brand lift from TikTok isn't captured." "Google ROAS dropped because Performance Max isn't giving us visibility."
These may all be true. They may all be rationalizations. The MER number is what it is. It went up or it went down. Investigate the platform numbers as diagnostics, but do not let the diagnostic story override what MER is telling you. If MER is down three months in a row, the business is becoming less efficient, regardless of what the platform ROAS explanations say.
Should I include email revenue in my MER denominator?
Include Klaviyo/Attentive platform fees in spend. Email revenue is already captured in total revenue in the numerator. The question is whether you attribute email-driven purchases separately from paid social purchases. For MER, you do not need to. The blended formula does not care which channel the conversion came through.
What if I have a high-LTV subscription business?
For subscription businesses, calculate MER on first-order revenue and track LTV:CAC separately. Using total lifetime revenue in the MER numerator overstates current-period efficiency because you are counting future revenue against current spend.
Does MER work for very small brands under $500K annual revenue?
Yes, but the monthly variance is higher because sample sizes are smaller. Use 3-month rolling MER instead of spot monthly at smaller scales.
Related reading
This piece is part of the paid social for DTC operators hub. The structural reason agencies rarely lead with MER is why most agencies overspend on top of funnel. For the measurement stack that makes MER diagnostics trustworthy, the Attribution Teardown reviews the event map and platform reports together.