ROAS is Misleading You & Killing Your eCommerce Business
Why The Most Quoted Number in Your DTC Business Is The Number That Lies Loudest.
The agency hands you the monthly report.
"4.2x ROAS this month. Up from 3.8x in March."
You smile. Then your bookkeeper, sitting in the next chair, pulls up the bank account.
It is $40K thinner than it was 30 days ago.
Both numbers are true. Both can be true at the same time. That is the most expensive lesson in eCommerce: ROAS is the most quoted metric in your business, and the most misleading number you will ever look at.
The elephant in the room
If ROAS is so misleading, why do even the most successful eight- and nine-figure DTC brands that build in public talk about it?
Open X on any given day.
The founders posting their daily ad-account screenshots, the operators sharing weekly ROAS deltas, the agencies posting case studies, almost all of them lead with ROAS.
You see this enough times and you start to assume that is how the winners run.
So you do the same. Daily ROAS check. Kill ads at 2.0x. Pour budget into anything above 4.0x. Build your weekly review around the ROAS column.
Here is what most brands miss when they copy that playbook.
Those few brands posting ROAS publicly are run by some of the most sophisticated DTC marketers in the industry. Their ROAS is not just the raw Meta number.
It is either calibrated through mature measurement methodologies like incrementality testing, geo-lift studies, and media-mix modeling, or validated against a robust business-KPI stack that most other brands do not have in place:
real-time contribution-profit dashboards,
cash-flow visibility,
channel-attribution validation against orders,
sometimes all of the above.
You do not have any of that. Most brands do not have any of that.
When a random DTC brand starts obsessing over ROAS without those foundations in place, ROAS stops being a metric and becomes the single biggest blocker to growth potential.
I dig deeper into this topic in the following article:
eCommerce brand with solid products, loyal customers stuck under $90K/m?
You built something real. Sales are coming in - maybe $15K last month, $20K the month before. You have a product people actually want, a Shopify store that functions, ads running, and content going out. By any reasonable measure, you are executing.
ROAS obsession tells you
to kill profitable campaigns
to cut ad spend at the worst possible time
the ad account is the problem when the actual problem is upstream, in your unit economics or your cash position, not in the ad account at all.
I have spent more than 10 years leading marketing and growth for DTC brands. The argument I have been making to founders for most of that time has not changed:
Rely less on ROAS, more on the actual indicators of business health.
eCommerce Finance Simplified to Help You Make More this year
eCommerce owners with better mastery over their finances have significantly higher net margins, more money in the bank, faster income growth, and pull capital out of their businesses at higher rates. (According to The 2026 eCom Trends Report by ECOM FUEL
This article is long because the lie inside ROAS is layered.
By the end, you will know exactly what ROAS measures, what it does not measure, and the three numbers that actually tell you whether your business is healthy.
What founders think ROAS means. What it actually says.
ROAS stands for Return on Ad Spend. The math is one line:
Revenue from ads divided by ad spend.
If you spend $1,000 and Meta says it generated $4,000, your ROAS is 4.0x.
What founders hear when they see "4.0x ROAS": every dollar I put into ads, I get four back. The ads are working. The business is healthy.
What ROAS actually says: Meta is claiming credit for $4,000 in attributed conversions.
Some of those were caused by the ad. Some would have happened anyway. Some are double-counted with another platform. None of it tells you whether you made money.
This is a hard sentence to swallow if you have been running on this number for years. So let us pull it apart.
Blended ROAS/ MER is better, but still misleading
Many advertisers use blended ROAS (Net Sales / Total Ad Spend) or MER (Total Ad Spend / Net Sales, expressed as a percentage) instead of platform-reported ROAS. They are better indicators of efficiency than platform-reported ROAS numbers, but they can also be misleading.
How eCommerce ROAS misleads D2C Founders
Here are a few examples from the countless ones I have seen over the last decade.
Same (Blended) ROAS, Opposite Outcomes
2 brands. Both are doing $30K/month in revenue. Both are spending $5K/month on ads. Both reporting 6.0x (Blended) ROAS.
Brand A:
COD (Cost of Delivery): 50% of net sales
Fulfillment overhead: $5K/month
OPEX: $7K/month
Net Profit: -$2K (loss)
Brand B:
COD: 30%
Fulfillment overhead: $3K/month
OPEX: $3K/month
Net Profit: $10K (profit)
Same revenue. Same ad spend. Same (blended) ROAS. One brand is losing $2K every month. The other is making $10K every month.
A 12-month gap of $144K between them, on identical (ROAS-based) ad performance.
ROAS does not see the cost structure. It compares revenue to ad spend and stops there. Whether you have a healthy business behind that revenue or a slow bleed, ROAS reports the same number.
(Quick definition: COD is your full Cost of Delivery, which is everything it actually costs to fulfill an order. Landed COGS plus receiving labor plus pick and pack plus packaging plus outbound shipping plus payment processor fees. If you only count base COGS and the carrier label, you are overstating gross profit. The gap between “COGS” and full COD is where most brands misread their unit economics.)
A higher ROAS at a lower scale can result in lower profit
A real example.
Assume a brand with 40% Cost of Delivery (COD) on its product and $900/day in fixed overhead (salaries, software, fulfillment infrastructure).
At $1,000/day in ad spend and 4.0x ROAS, the brand is doing $4,000/day in revenue. Subtract COD ($1,600), ad spend ($1,000), and overhead ($900). Net Profit: $500/day.
They increase ad spend to $5,000/day. ROAS drops to 2.5x. Revenue climbs to $12,500/day. Subtract COD ($5,000), ad spend ($5,000), and overhead ($900). Net Profit: $1,600/day.
Lower ROAS. More than three times the profit.
Another example, this brand has $2M+ in monthly revenue on Amazon (that means they have inventory and capital), but their Shopify revenue remains below $100K per month, while they worry about protecting their 9+ blended ROAS. At higher spend levels, even with projected lower blended ROAS, they could generate more contribution profit, operating at a higher revenue stage.
Most founders look at the ROAS drop and panic. They cut the budget back, the higher ROAS returns, and the brand stays stuck in a perpetual low-revenue cycle.
I have watched this play out at countless brands. Some of them with great products. Real PMF. Big enough TAM. They locked themselves in by chasing the metric.
Higher ROAS is not the same thing as more profit.
The math gets clearer as you scale, but the founder mindset gets harder. You have to be willing to watch your ROAS go down on purpose.
The “cut spend to fix ROAS” death spiral
When ROAS looks bad, the instinctive move is to cut ad spend. This is the move that quietly kills more bootstrapped brands than any creative problem.
I wrote a longer piece on this:
The Hidden Cost of Cutting Ad Spend: Why Your P&L Says Profit But Your Bank Account Says Panic
You’re staring at your dashboard at 11 PM. Your profit margins are barely in the double digits. Your MER is sitting at an ugly 56%. You’re spending more than half your revenue on ads just to keep the lights on. Your CFO (or accountant, or that voice in your head) is saying the same thing everyone says: “We need to get more efficient. Cut the ad spend. F…
The summary: cash velocity matters more than margin at scale, and ROAS does not see cash velocity at all.
Cold customer acquisition is costly through ads, but it isn’t optional unless you can build an alternative channel that can steadily feed the funnel with new audiences and potential customers.
If you cut budget every time ROAS dips, you are not optimizing. You are starving the system.
ROAS is blind to what happens after the first sale
Two campaigns:
Campaign A: 3.0x ROAS, brings in one-time buyers.
Campaign B: 2.0x ROAS, brings in customers who make 5 repeat purchases.
ROAS thinks Campaign A is the better one.
LTV math thinks Campaign B is way better.
Most founders run on a metric that is structurally blind to the most important question in eCommerce: who comes back, how often, and at what AOV.
Attaching some numbers shared by the operators building in public of some of the fastest‑growing DTC brands that had the LTV play. I hope this helps you to set more realistic expectations.
Blended ROAS, MER, or CPA doesn’t tell the whole story.
When you see the above screenshot (from a 7‑figure brand), that this brand was getting 12+ blended ROAS, or they were just spending 8% of their net sales into advertising, what comes to your mind?
Their advertising is going great, and the brand can be scaled to the moon, right?
But that’s not true at all.
This brand has scaled to a 7‑figure business purely based on category demand over a decade.
Their advertising spend was below $7K/m and contributed insignificantly to their eCommerce revenue.
As long as the rising tide of the category demand drove their business, their unit economics worked.
Their world‑class supply chain and operations served countless happy customers.
But as their natural category demand started declining and their revenue and profit numbers started going down year after year, they could not just throw money into advertising to bounce back profitably.
The usual media buying playbooks of ad-driven eCommerce brands don’t apply to brands like this one.
You need to account for the organic baseline, which, though inconsistent, still drives the majority of revenue. Ensuring that ad spend actually yields incremental revenue and doesn’t merely steal credit for purchases that would have happened anyway, thereby eroding margins, is extremely complicated.
Those ROAS and CPA metrics can’t help you with that.
“Show me a successful brand that grew without high ROAS.”
Take Chubbies: a nine-figure, expanding omnichannel business. Eight years of growth in both the top and bottom lines, with 40%+ EBITDA growth year over year.
Good enough?
1X ROAS on Meta vs 10X ROAS on Google - where to spend more?
I was auditing a brand that was getting under 1.0x ROAS on Meta and TikTok and 10x or higher on Google.
Their decision: kill Meta and TikTok, scale Google.
Within weeks, everything crashed.
What I pointed out to them is the same thing I’ve been showing to e-commerce brands for over a decade:
The Meta and TikTok spend had been creating demand, while Google was converting the demand and stealing credit at the last click. Once the demand engine was off, there was nothing left for Google to harvest.
As the founder, you need to understand how different ad platforms, marketing channels, campaign types, and content types work.
Google’s branded search campaigns, channels like direct and organic search, can be heavily influenced by platforms and strategies that drive more brand awareness and get discovered by a net‑new audience - Like video content, both organic and paid, on social commerce platforms like Meta and TikTok.
The highest ROAS ads/campaigns/platforms might NOT be the top-performers
Meta & most ad platforms use last-click attribution by default.
The ads showing the highest ROAS are typically retargeting and bottom-of-funnel. They often get credit for sales that were also influenced by other upper-funnel ads from the same platform, other ad platforms, or other marketing touchpoints, or all of them.
In most cases, those ads claim sole credit for a conversion that was influenced by other touch points, and in some cases, they claim credit for a conversion that would have happened anyway, even if those customers hadn’t seen those ads.
You can extend the same idea to ad sets, campaigns, and ad platforms as well.
You cannot rely on platform ROAS (alone) as a measurement of what works.
Tracking was never accurate, and post-iOS 14, it is meaningfully worse.
Third-party attribution platforms claim to solve this, but more advertisers are waking up to the misleading nature of MTA (multi-touch attribution) or any click-based measurement system alone.
A full budget allocation decision‑making framework is beyond the scope of this article.
But to simplify, the core idea is as follows:
The ultimate objective of running advertisements on one platform or multiple platforms is to improve your business’s financial health today and tomorrow.
So whenever we are evaluating any ad platforms, campaigns, or ads, our evaluation framework needs to be tied to our actual desired business results, not any vanity metrics.
Why CPMr & Reach are buzzing on DTC X Community
I’m so glad to see interactions happening around topics like the one attached below.
If you aren’t spending six figures monthly, you may not realize how important this lens is. But if you’ve been spending six figures on ads for a while and have hit a ceiling where you can’t scale further, you should definitely audit this in your ad accounts.
At a higher budget, you are going to run out of warmed‑up in‑market audience much faster. And you need a constant supply of new audience getting fed into your funnel.
High-ROAS ads tend to have high CPMr (cost per 1,000 accounts reached), which means the algorithm is serving them to a small warm audience repeatedly and claiming credit for conversions that were already in motion.
The ads doing the real acquisition work, reaching net-new people at scale, look terrible on paper because attribution rewards the harvest and almost never rewards the planting.
You can verify this in your own account in 10 minutes:
Sort your top 10 ads by spend, add CPMr and Reach columns,
then look at your top 3 ads by ROAS and your top 3 by Reach.
If those are different ads, you have the pattern.
Ditch ROAS, Obsess on Revenue + Profit (& Cash) Instead
I couldn’t express this more simply than Nate Lagos did in this podcast episode, the former VP of Marketing at Original Grain.
If you take one thing from this article, take this list. These are not platform metrics. They are the actual indicators of business health.
1. Net Sales.
Revenue after discounts and returns.
The only valid base for profitability math.
Most dashboards default to "Total Sales" which includes shipping fees and tax.
Both are pass-throughs. Strip them out. What is left is your actual product revenue.
2. Contribution Profit dollars.
Net Sales minus Total COD minus Total Ad Spend.
This is what actually pays your overhead, salaries, software stack, and rent. It is the number that decides whether the business has fuel to scale or is about to start eating itself.
Pay attention to the word "dollars." Not "margin %."
A 30% contribution margin on $150K is $45K. A 50% contribution margin on $30K is $15K. The first business has more money to invest in growth, in inventory, in talent. Ratios can mislead. Dollars cannot.
3. Cash position and cash conversion cycle.
Cash position is what is in the bank, less what you owe in the next 30, 60, 90 days.
Cash Conversion Cycle (CCC) is how long your cash is trapped between paying suppliers and getting paid by customers.
The formula: DIO + DSO - DPO.
Days Inventory Outstanding plus Days Sales Outstanding minus Days Payable Outstanding.
Most DTC brands have a CCC of 30 to 60 days. That means every dollar of growth needs another 30 to 60 days of cash to fund it.
If you are chasing higher ROAS by cutting volume, your CCC gets worse, not better. Inventory turns slower. Cash gets more trapped.
Together, these three numbers tell you what ROAS cannot:
Are we actually making money on the orders we run? (Contribution Profit dollars)
Can we afford to keep doing it? (Cash position + CCC)
Are we growing the base we are making it on? (Net Sales)
Business health is the combination of those three. Not a ratio.
Are you building a brand or doing ad arbitrage?
This is the question hiding underneath every "is my ROAS good?" conversation, and almost no one asks it out loud.
I’m so glad that Preston Rutherford keeps talking about this & delivered this session at the Meta Performance Marketing Summit 2026.
Think about all the legacy brands that stood the test of time before this digital ad platform era, the ROAS-obsessed era of performance marketing. How did they grow with traditional marketing and advertising when they could not evaluate ROAS from the campaigns they ran?
They understood this, and they focused on building a brand.
Many digital‑first founders and advertisers assume legacy enterprise brands spend on marketing without a scientific measurement methodology. Nothing could be farther from the truth. As you scale, go omni‑channel, and adopt more mature measurement approaches, you’ll find they are far more scientific than click‑based methods like attribution.
Obsess on ‘Incrementality’ if you are an 8‑figure and omnichannel brand
Before that stage, you can focus on your revenue, profit, and cash for the most impactful decision-making.
Past that scale, or when your channels (both marketing and sales) start interacting in ways the spreadsheet cannot capture, the next measurement layer is real:
Incrementality testing (geo-lift, holdouts, conversion-lift studies).
Marketing Mix Modeling (MMM), calibrated.
Multi-Touch Attribution (MTA), calibrated.
The right way to think about these layers is not "pick one." It is "use lift-based calibration to bridge MMM and MTA."
MMM gives you the broad allocation view with low precision but cross-channel comparability.
MTA gives you high-precision touchpoint data but weak causality.
Lift studies give you episodic ground truth. The lift studies calibrate the other two.
That is a topic for another article.
Run the math you should have been running
Here is the math most founders never learn. It takes two minutes. It is the difference between killing profitable campaigns and scaling them.
Step 1: Calculate your Gross Margin.
Net Sales minus Total COD, divided by Net Sales.
e.g. Selling price $58. Total COD $29.10. Gross Margin = ($58 - $29.10) / $58 = 49.8%.
Step 2: Calculate your break-even ROAS.
1 divided by Gross Margin %.
From the above example: 1 / 0.498 = 2.01x.
This is your floor. Below 2.01x (blended) ROAS, every dollar of ad spend is producing less than a dollar of gross profit. You are paying to fulfill orders.
Above 2.01x, every dollar of ad spend is producing more gross profit, covering the ad spend and generating contribution profit.
I have watched countless brands kill campaigns at 2.1x for years.
The campaigns were profitable. The founders just did not know what number to compare them to.
This is the calculation every founder should do once and write on the whiteboard.
Your break-even ROAS is a function of your gross margin, not a function of "industry benchmarks."
The 2x2 operating system
For brands not yet at an 8-figure omnichannel scale (where MMM and incrementality testing become the right investment), here is the operating system I run with my clients. Two report types, two cadences.
The 2x2:
Business financials (daily and monthly)
Meta account-level reports (daily and monthly)
Report 1: Business financials
The P&L waterfall:
Net Sales
Total Cost of Delivery (COD)
Gross Profit (Net Sales minus COD)
Total Ad Spend
Contribution Profit (Gross Profit minus Ad Spend)
That is it. Five lines. Every day, every month.
The goal: increasing Net Sales and increasing Contribution Profit, month over month. That is the game.
Notice what is not in this report: ROAS. Not because it is useless. We use it inside the Meta report. But because it is not the indicator of whether the business is winning. Contribution Profit dollars are.
Report 2: Meta account-level
The metrics that matter:
Amount spent
Cost per Purchase (CPP / CPA)
Average Purchase Conversion Value (AOV)
AP (Ad Profit) = Purchase Conversion Value minus Amount Spent. Proxy for contribution profit inside the platform. You do not have access to real Gross Profit or Contribution Profit inside Meta, so AP is the closest thing.
APT (Ad Profit per Transaction) = AOV minus CPA. The per-order profitability proxy. APT is more useful than AP when comparing periods at different spend levels, because AP has volume bias.
ROAS, used as an auto-rule trigger, not as a primary diagnostic.
CPM, CPMr (Cost per 1,000 Accounts Reached), Frequency, as supporting signals.
How they connect
When AP and APT are positive and trending up in the daily Meta report, and the backend financial report confirms Contribution Profit is growing, that is the scaling signal.
When AP looks fine but backend Contribution Profit is not growing, the discrepancy needs to be investigated before you scale.
Scale because the 2x2 confirms it. Not because Meta recommends it. Not because your MTA tool shows good ROAS. Because both layers, platform proxy and backend reality, are pointing the same direction.
When ROAS is actually useful
ROAS is not garbage. It is just badly used. Here is when it is genuinely useful, with the right calibration.
As an auto-rule trigger.
(Specifically, if you don’t want to use manual bids and are using auto bids, the highest volume or the highest value.)
Meta's auto rules support ROAS as a condition. They do not support AP or Contribution Profit. So if you want to automate budget adjustments, increase budget when performance is strong, decrease when it weakens, ROAS is a decent available trigger.
This works when you derive the ROAS threshold from your own data, not from "industry benchmarks":
Export daily Meta account-level ROAS for the past 90 days.
Export daily backend Contribution Profit for the same period.
Find the ROAS threshold above which you have CP-positive days, and below which you have CP-negative days.
Use that range as your starting auto-rule threshold.
Tune it over time. If backend CP is healthy and growing, lower the threshold (2.0 then 1.9 then 1.8) and raise the budget cap. If backend CP is weak, raise the threshold and lower the cap.
A simpler founder rule (if all of this feels like too much)
If you do not have time to build a 2x2, do not have a finance person, and want one rule to operate on, use this:
If your Contribution Profit dollars are growing month over month, your engine is working. If they are not, no ratio will save you.
That is the entire rule.
Net Sales up + Contribution Profit dollars up = healthy growth. Keep going.
Net Sales up + Contribution Profit dollars flat or down = you are buying revenue. Stop and assess your business model and access to capital. If acquiring more customers and capturing more of the market, even at lower contribution profit dollars, still makes sense.
Net Sales down + Contribution Profit dollars up = you are over-cutting. Probably starving acquisition. Watch the next cycle carefully.
Net Sales down + Contribution Profit dollars down = something structural is wrong. Diagnose at the financial layer first, not the ad account.
ROAS does not appear in any of these decisions. Not because it is invisible. Because it is downstream of the things that actually matter.
I am not a finance person. I am a marketer who was forced to learn the financial side because too many founders I worked with were running profitable-on-paper businesses straight into bankruptcy by chasing the wrong number. ROAS was the wrong number more often than any other metric.
I still see this every week.
A founder kills a profitable campaign because ROAS is "low."
A bookkeeper who does not understand marketing tells the founder to cut spend "to fix the numbers."
An agency reports a strong ROAS in a deck, and the brand quietly burns down.
Here is the question I want you to sit with after reading this:
If you stopped looking at ROAS for the next 90 days and made every ad budget decision from your Contribution Profit dollars and cash position alone, what would change?
For most brands I have worked with, the honest answer is: they would scale. The ones already scaling would scale faster. The ones stuck would unstuck.
Try it. Pull the ROAS column out of your weekly review for one quarter. Replace it with Contribution Profit dollars and cash position. Watch what happens to your decisions. Watch what happens to your business.
I am still learning the financial side of this every year. But I have not made an ad spend decision based on ROAS alone in a long time, and neither should you.