What is NMER?

By the HonestTag team · Published July 11, 2026

MER became popular for a good reason: when platform attribution stopped being trustworthy, dividing total revenue by total spend was at least arithmetic nobody could argue with. But blended MER answers a fuzzy question. NMER sharpens it into the question an acquisition budget actually has to answer.

The formula

Over any period (a month is typical):

• Blended MER = total revenue / total ad spend
• NMER = revenue from new customers / total ad spend

The numerator is the only difference, and it changes everything. Total revenue includes your repeat buyers, your email flows, your word-of-mouth, and everyone who typed your brand into Google. New-customer revenue is the part your acquisition engine can plausibly claim to have created.

A worked example

A Shopify store does $100,000 in revenue this month on $20,000 of ad spend. Blended MER: 5.0. Feels comfortable.

Split the revenue: $65,000 came from returning customers and email, $35,000 from first-time buyers. NMER: 35,000 / 20,000 = 1.75.

Now the same store six months ago: $80,000 revenue, $15,000 spend, $40,000 of it new-customer. Blended MER was 5.3, NMER was 2.7. Blended MER barely moved (5.3 to 5.0) while NMER fell from 2.7 to 1.75, a 35% collapse in acquisition efficiency, masked by a growing base of repeat buyers that this month's ads did not earn. Stores discover this drift months late when they only watch the blended number, usually at the moment repeat revenue plateaus and there is no acquisition engine left under it.

Why blended MER flatters your ads

Repeat revenue is a lagging asset. It was created by past acquisition, product quality, and retention work, but it lands in the numerator of blended MER every month regardless of what your current ads are doing. The stronger your retention gets, the more your blended MER inflates, and the easier it is to keep funding mediocre acquisition. It is the same self-grading problem as platform attribution, one level up: the metric borrows credit from work the spend did not do. HonestTag's position is simple: metrics should not borrow credit. That principle is why NMER exists, and why it pairs with NCAC, its cost-side twin.

What you need to compute it honestly

The formula is trivial; the numerator is not. "New customer" has to mean verified-new against your store's order history, not a platform checkbox. Getting it right takes three inputs:

1. Accurate new-vs-returning classification per order, from your own Shopify data. Guest checkouts and email variants make this harder than it sounds.
2. All ad spend in one place, across platforms, including the ones without APIs (that is why HonestTag supports manual spend lines alongside connected sources).
3. Refund handling. A refunded first order is not acquired revenue, so the store-level NMER numerator is reduced. Google Ads and Microsoft Advertising receive the available retraction or restatement, and Klaviyo and Google Analytics 4 receive a refund event. HonestTag does not send Meta, TikTok, Pinterest, Snapchat, Reddit or OpenAI Ads a refund or retraction event; the refund is recorded in order proof. Channel revenue in Verdicts is first-click new-customer revenue, gross of refunds. Refunds carry no per-channel split, so netting them there would be invented precision; store-level MER and NMER stay net. See first-party tracking for how the underlying measurement works.

Judging NMER while you scale

NMER falls as spend rises. That is not failure; it is arithmetic. New spend buys the next-cheapest customers, so efficiency per dollar dilutes while total acquisition grows. The pair to judge a scale-up by is NMER times new-customer count: if that product grew while spend rose, the scale-up is working even though NMER dipped. HonestTag bakes this into its weekly verdicts: a channel is never marked Cut just because NMER decayed under rising spend, and the verdict names the decay when it happens. For the wider MER family and when to use each ratio, see the MER guide.

The honest tradeoffs

• NMER ignores lifetime value. It scores the first order only. A subscription store can happily run NMER below 1.0 because month two makes the math work; a one-and-done category cannot. Read NMER next to LTV and payback, not instead of them.
• It is account-level arithmetic. NMER will tell you acquisition efficiency is decaying, but not which campaign is at fault. For that you need per-channel attribution underneath, ideally first-click on new customers, so the account-level and campaign-level stories reconcile.
• Organic new customers ride along. Some first-time buyers would have arrived with zero spend. NMER cannot separate them; only incrementality testing can. Treat NMER as an honest ceiling on paid efficiency, not proof of causation.

Frequently asked questions