D2C Unit Economics in 2026: CAC, LTV, Contribution Margin, and the Math of Profitable Growth

D2C unit economics is the math of one order and one customer: what you collect, what it costs to make, ship, and sell, and what is left to pay for ads, overhead, and profit. If that number is negative, growth makes you poorer. If it is positive, growth is a funding question, not a survival question.

Here is the short answer most founders are looking for.

• You are profitable on an order when ad cost per order is lower than contribution margin per order. Contribution margin is revenue minus product cost, shipping, payment fees, pick and pack, and returns.
• Break-even ROAS is 1 divided by your contribution margin percentage. At a 50% margin you need a 2.0 ROAS to break even. At 30% you need 3.33.
• The median Meta advertiser is close to the line. Across Triple Whale's ecommerce benchmark set (August 2025 to July 2026), median Meta ROAS was 1.88 and median cost per acquisition was $38.99. A 1.88 ROAS only breaks even on the first order if your margin after variable costs is above 53%.
• Judge the business on blended numbers, not platform dashboards. Marketing efficiency ratio (total revenue divided by total ad spend) and new-customer CAC are harder to game than platform ROAS.
• Measure LTV over 60, 90, and 180 days from real cohorts. Do not project three years forward from four months of data.

The rest of this guide shows the formulas, the public benchmarks behind each claim, and worked examples you can copy into a spreadsheet. Every statistic is linked to its source and listed again at the end. If you would rather have someone run the numbers and the campaigns with you, that is what our Meta Ads management and Google Ads management work covers. Everything below, you can do yourself.

What are D2C unit economics?

Unit economics answers one question: does each additional order, and each additional customer, add money to the business or take it away?

There are two units worth tracking.

• The order. Revenue per order minus every cost that scales with that order. This gives you contribution margin.
• The customer. What it cost to acquire them (CAC) against the contribution margin they generate over time (LTV).

Fixed costs like salaries, software, and rent are left out on purpose. They matter for the company P&L. They do not tell you whether selling one more unit is a good idea. Unit economics does.

A brand with good unit economics and high fixed costs has a scale problem. A brand with bad unit economics has a model problem.

How do you calculate profit on a single order?

Build an order-level P&L. Start with what the customer pays and subtract each variable cost in layers. The layers are usually called contribution margin 1, 2, and 3.

There is no accounting standard for these labels. Finance teams draw the lines in slightly different places. The definitions below are the most common operator convention, and what matters is that you pick one and keep it fixed.

Layer	Formula	What it tells you
Net revenue	List price minus discounts, excluding sales tax or GST you collect	What you actually keep from the sale before costs
Contribution margin 1 (CM1)	Net revenue minus landed cost of goods	Product margin. Is the product itself viable?
Contribution margin 2 (CM2)	CM1 minus shipping, pick and pack, packaging, payment fees, and a returns allowance	Margin after delivering the order. This is the number ads must beat
Contribution margin 3 (CM3)	CM2 minus ad spend per order	What one order contributes to fixed costs and profit

A worked example: one $60 order

Take a skincare brand with a $60 average order at list price. New customers use a 10% welcome code. The brand pays for shipping. Payments run through a standard 2.9% plus 30 cents card rate. The numbers are illustrative, the structure is what you copy.

Line	Amount	% of net revenue
List price	$60.00
Discount (10% welcome code)	-$6.00
Net revenue	$54.00	100.0%
Landed cost of goods	-$15.00	27.8%
CM1	$39.00	72.2%
Outbound shipping	-$6.50	12.0%
Pick, pack, and packaging	-$2.50	4.6%
Payment fee (2.9% of $54 + $0.30)	-$1.87	3.5%
Returns allowance (5% of orders at about $30 each)	-$1.50	2.8%
CM2	$26.63	49.3%
Ad spend per order	-$22.00	40.7%
CM3	$4.63	8.6%

Three things stand out.

First, a product with a 72% product margin ends up with 49% after delivery. Shipping, fees, and returns took 23 points. Founders who quote "70% margins" and plan ad budgets on that number overspend.

Second, the $6 discount came straight out of margin. It was 10% of list price and 18% of CM2 before the discount (6 divided by 32.63).

Third, CM3 of $4.63 is what is left to cover salaries, software, content, and profit. At 1,000 orders a month that is $4,630.

What goes in landed cost of goods?

Landed cost is what a unit costs you sitting in your warehouse, ready to ship. Include the factory price, inbound freight, duties and tariffs, and any inspection or prep fees. Tariffs are a live issue. Warby Parker reported gross margin of 54.0% for 2025, down from 55.3% a year earlier (FY2025 10-K), and Allbirds tied its own decline partly to higher duties. If you import, recompute landed cost every time a rate changes.

What do payment processors actually charge?

Payment fees are small per order and large per year. These are the published standard rates. Check the vendor's page before you model, because rates vary by country and plan.

Processor	Published standard online rate	Notes
Shopify Payments (US, Basic plan)	2.9% + $0.30	Grow plan 2.7% + $0.30. Advanced plan 2.5% + $0.30
Shopify with a third-party gateway	Adds 2% (Basic), 1% (Grow), or 0.6% (Advanced) on top of the gateway's own fee	Waived when you use Shopify Payments
Stripe (US)	2.9% + $0.30	Plus 1.5% for international cards and 1% for currency conversion
PayPal Checkout (US)	3.49% + $0.49	Standard card payments 2.99% + $0.49
Razorpay (India)	2% platform fee plus 18% GST on the fee, about 2.36% in total	International cards 3% plus GST, about 3.54% in total

Sources: Shopify pricing, Stripe pricing, PayPal business fees, Razorpay pricing. Rates as published in 2026.

The fixed fee matters more on small orders. On a $15 order, 2.9% plus 30 cents is 4.9% of revenue. On a $150 order it is 3.1%. Low-AOV brands pay a structurally higher fee rate.

How should you account for returns?

Returns are not an edge case. The NRF and Happy Returns 2025 Retail Returns Landscape report, based on a summer 2025 survey of 358 ecommerce professionals at large US retailers and 2,006 consumers, estimates that 15.8% of annual retail sales will be returned in 2025, about $849.9 billion. For online sales the estimate is 19.3%.

Those are large-retailer, all-category figures. Apparel and footwear run higher. Consumables run far lower. Use your own rate, and model it as an allowance on every order:

Returns allowance per order = return rate x cost of one return

The cost of one return is the margin you refund, plus return shipping if you pay it, plus handling, plus the write-down if the item cannot be resold at full price. In the example above, 5% of orders come back at roughly $30 each, so every order carries $1.50.

A return also wastes the ad spend that produced the order. If 19.3% of orders came back, your CAC per kept order would be 24% higher than the dashboard shows (1 divided by 0.807).

What gross margin should a D2C brand have?

Enough that CM2 can pay for acquisition with something left over. In practice that means product margin matters more than almost any marketing decision you will make.

There is no audited benchmark of private D2C brand margins. The best public evidence is what listed, digitally native brands report. These are gross margins from their own filings and earnings releases.

Company	Category	Reported gross margin	Period	Source
Hims & Hers Health	Telehealth and personal care	74%	FY2025	Form 10-K
FIGS	Apparel (healthcare)	66.5%	FY2025	Earnings release
YETI	Drinkware and outdoor	57.4% (adjusted)	FY2025	Proxy statement
Warby Parker	Eyewear	54.0%	FY2025	Form 10-K
Allbirds	Footwear	41.0%	FY2025	Form 10-K

Read this table with two cautions. Public companies define gross margin differently. Some include fulfillment and shipping in cost of goods and some do not, so these are not CM2 figures. And these are scaled businesses with buying power you do not have yet.

The pattern still holds. Beauty and personal care sit around 70%. Branded apparel and accessories sit in the mid 50s to mid 60s. Footwear is lower. Reselling other people's products sits near 30%, which is why resellers depend on repeat purchases and subscriptions rather than first-order profit.

If your product margin (CM1) is under 50%, paid acquisition on the first order will be very hard to make profitable. The break-even table below shows why.

What ROAS do you need to break even?

Break-even ROAS is the return on ad spend at which ad cost exactly equals contribution margin. Below it, each order loses money before fixed costs. The formula is short.

Break-even ROAS = 1 / CM2 margin %

Break-even CAC (first order) = CM2 per order in dollars

In the worked example, CM2 is 49.3% of net revenue. Break-even ROAS is 1 divided by 0.493, which is 2.03. Break-even CAC is $26.63. The brand is paying $22, so it clears the bar with $4.63 to spare.

Break-even ROAS lookup table

CM2 margin (after product, shipping, fees, returns)	Break-even ROAS	Max ad cost as % of revenue	Break-even CAC on a $60 net order
20%	5.00	20%	$12.00
25%	4.00	25%	$15.00
30%	3.33	30%	$18.00
35%	2.86	35%	$21.00
40%	2.50	40%	$24.00
45%	2.22	45%	$27.00
50%	2.00	50%	$30.00
55%	1.82	55%	$33.00
60%	1.67	60%	$36.00
70%	1.43	70%	$42.00

Use CM2, not product margin, in this table. That is the most common mistake in break-even math. A brand with a 70% product margin and a 49% CM2 has a break-even ROAS of 2.03, not 1.43.

Make sure the ROAS you compare uses the same revenue definition. If your margin is calculated on net revenue after discounts and excluding tax, your ROAS must be too. Platform dashboards often report gross order value including shipping and tax, which flatters the number.

Why the median Meta account sits near break-even

Triple Whale's Meta benchmark for August 2025 to July 2026 shows these medians across its ecommerce customer base.

Metric (Meta ads, all ecommerce)	Median	Detail
ROAS	1.88	Home and Garden 2.25. Apparel and Accessories 2.24. Health and Wellness 1.44
Cost per acquisition (CPA)	$38.99	Apparel and Accessories $36.98. Home and Garden $47.93. Electronics $51.86
CPM	$15.06	Up about 13% year over year
Click-through rate	2.39%
Conversion rate	1.53%
Average order value	$73.36

Source: Triple Whale, Facebook Ad Benchmarks by Industry, data window August 2025 to July 2026. Platform-attributed figures from brands using Triple Whale.

A median ROAS of 1.88 means half of advertisers are below it. To break even at 1.88, you need a CM2 margin of 53.2% (1 divided by 1.88). Most categories in the gross margin table above do not have that after shipping and fees.

So the typical brand is not making money on the first Meta-attributed order. It is buying a customer and betting on what happens next. That only works if you have measured what happens next.

Should the first order be profitable, or is payback enough?

There are two acquisition models. Know which one you are running.

First-order profitable. CAC is lower than first-order CM2. Every new customer pays for themselves immediately. Growth is self-funding and limited only by how much volume exists at that CAC. This is the right default for a young brand with little cash and no retention data.

Payback period. CAC is higher than first-order CM2. You lose money on order one and earn it back from repeat orders. Growth consumes cash until the cohort pays back.

Payback period = the number of days until cumulative contribution margin from a cohort equals what you paid to acquire it.

Payback is not wrong. Subscription and replenishment brands run on it. But it needs proof that customers come back, and enough cash to fund the gap.

A simple rule: your payback window should be shorter than the time your cash can cover. If you have three months of runway for marketing, a 180-day payback will run you dry while the spreadsheet says you are winning.

What is a good CAC for a D2C brand?

A good CAC is one that is lower than the contribution margin the customer generates within a period you can afford to wait. There is no universal dollar figure. A $60 CAC is excellent for a $400 mattress topper with 55% margins and fatal for a $28 candle.

Benchmarks are useful as context for what the market charges, not as targets.

Benchmark	Figure	Data and method	Source
Median Meta CPA, all ecommerce	$38.99	Triple Whale customers, August 2025 to July 2026, platform-attributed	Triple Whale
Median Meta CPM, all ecommerce	$15.06, up about 13% year over year	Same dataset	Triple Whale
Google Search average CPC, all industries	$5.42 (2026 edition), up from $5.26 (2025 edition)	US Google and Microsoft search campaigns, April 2025 to March 2026	WordStream by LocaliQ
Facebook traffic campaigns	CPC $0.60, click-through rate 1.93%	US campaigns, 2026 edition	WordStream by LocaliQ
Loss per new customer acquired	$29 in 2022, up from $9 in 2013	Vendor study, fully loaded cost including returns, all acquisition cost charged to the first order	SimplicityDX

Note that the WordStream figures cover all industries and lead generation as well as ecommerce, and a click is not a customer. Search clicks cost more than social clicks and convert better. Compare channels on cost per new customer, never on cost per click.

Is customer acquisition really getting more expensive?

The published evidence says yes, with caveats about who is publishing it.

• Auction prices are up. Triple Whale reports median Meta CPM up about 13% year over year to $15.06, while median CPA rose about 3%. WordStream's average Google Search CPC moved from $5.26 in its 2025 edition to $5.42 in its 2026 edition.
• Conversion partly offsets it. WordStream's 2026 edition reports the average search conversion rate at 8.18% across industries, which is why cost per lead in that report fell slightly while clicks got dearer. Costs per impression are rising faster than costs per customer.
• Longer-run studies point the same way. SimplicityDX reported in July 2022 that brands lost an average of $29 on each new customer, up from $9 in 2013, a 222% increase. ProfitWell (now Paddle) found CAC up roughly 60% over five years for both B2B and B2C subscription companies.

Treat the last two with care. SimplicityDX sells social commerce software and its figure assumes the first sale carries the full acquisition cost. The ProfitWell study covers subscription businesses, not physical products, and is several years old. They are directional evidence, not measurements of your market.

The practical conclusion does not depend on the exact number. Build your model so it still works if CAC rises 15%.

Blended CAC, paid CAC, and nCAC: which one should you use?

"CAC" is used for at least four different numbers. They can differ by a factor of two in the same business.

Metric	Formula	Use it for
Platform CPA	Platform ad spend / purchases the platform claims	Comparing ads and campaigns inside one platform. Not for profitability
Paid CAC	Total paid media spend / new customers attributed to paid	Judging the paid program as a whole
Blended CAC (also called nCAC)	Total marketing spend / all new customers in the period	Judging the business. The number for LTV to CAC
Fully loaded CAC	(Ad spend + agency, tools, creative, and marketing salaries) / new customers	Board-level and fundraising conversations

Platform CPA is the least trustworthy for profit decisions because the denominator includes returning customers and anyone who saw an ad and bought anyway. Triple Whale's own documentation defines new customer CPA as ad spend divided by orders from new customers, precisely to strip repeat buyers out.

For day-to-day decisions, track blended new-customer CAC. Shopify reports first-time customers by day. Divide total ad spend by that count. Nobody can dispute the inputs.

What is MER, and why do operators prefer it to platform ROAS?

Marketing efficiency ratio (MER) is total revenue divided by total ad spend across every channel, for the same period.

MER = total store revenue / total ad spend

aMER (acquisition MER) = new-customer revenue / total ad spend

Ad spend as % of revenue = 1 / MER

MER uses two numbers you can verify: what the bank received and what the ad platforms charged. No attribution model is involved. That is why operators trust it when platform dashboards disagree with each other.

One naming trap. Some tools invert the ratio. Triple Whale's MER definition is blended ad spend divided by order revenue, expressed as a percentage, with "blended ROAS" as the revenue-over-spend version. A MER of 25% in that convention is a MER of 4.0 in the common one. Check which way your dashboard runs.

A worked example: one month, three views of the same store

A store does $100,000 in net revenue in a month and spends $25,000 on ads.

Line	Meta	Google	Whole store
Ad spend	$18,000	$7,000	$25,000
Revenue the platform claims	$54,000	$35,000	$89,000 claimed in total
Platform ROAS	3.00	5.00
Actual net revenue			$100,000
Revenue from new customers			$55,000
Revenue from returning customers			$45,000
New customers			1,000
MER			4.00
aMER			2.20
Blended new-customer CAC			$25.00

The platforms together claim $89,000 of a $100,000 month. Yet $45,000 came from returning customers, many of whom would have bought from an email or by typing in the URL. Both platforms are counting some of the same orders, and both are taking credit for repeat buyers.

Platform ROAS says 3.0 and 5.0. MER says 4.0. The number that tells you whether acquisition is working is aMER at 2.2. If this store's break-even ROAS is 2.03, it is acquiring customers at a thin first-order profit, and the healthy-looking MER is being carried by the existing customer base.

Break-even aMER equals break-even ROAS. Run that comparison every week. A falling aMER with a steady MER means you are living off past customers.

MER has a weakness. It cannot tell you which channel or ad to change. Use MER and aMER to decide how much to spend. Use platform data to decide where inside the platform to spend it.

Which attribution numbers can you trust after iOS 14.5?

None of them completely. Each source is wrong in a known direction, so use them together.

Apple introduced App Tracking Transparency with iOS 14.5 in April 2021. Apps had to ask permission to track. Meta's CFO told investors in February 2022 that the change would cost the company on the order of $10 billion in 2022 revenue, as CNBC reported. According to Adjust, the industry-wide opt-in rate among users shown the prompt was 35% in Q2 2025. Most iPhone users are still not individually trackable across apps.

The platforms responded with modeled conversions. Reporting windows keep shifting too. Meta announced that from January 12, 2026 its Ads Insights API no longer returns 7-day view and 28-day view attribution windows. Year-over-year comparisons of platform ROAS can change because the reporting changed, not because performance did.

Source	What it is good for	How it misleads
Meta and Google dashboards	Comparing creatives, audiences, and campaigns within the platform	Over-credits. Counts view-through and returning customers. Uses modeled data
GA4	Seeing click paths and landing page behavior	Under-credits social. Misses view-through and cross-device. Data-driven attribution is the default model and is hard to audit
Post-purchase survey ("How did you hear about us?")	Finding channels clicks cannot see, such as podcasts, creators, and word of mouth	Self-reported. People name the last thing they remember
MER, aMER, and blended CAC	Deciding total budget and judging profitability	Cannot assign credit to a channel
Holdout or geo tests	Measuring true incremental lift	Slow, and needs volume most small brands do not have

The working rule: set budgets on blended numbers, optimize inside platforms on platform numbers, and use a one-question post-purchase survey as a sanity check. When they disagree, the bank account wins.

What is a good LTV to CAC ratio, and does 3:1 apply to ecommerce?

The 3:1 rule comes from software, not ecommerce. It is most often traced to David Skok's SaaS Metrics 2.0, which advises that lifetime value should be more than three times CAC and that CAC should be recovered in under 12 months. Those guidelines were written for subscription software with gross margins around 80% and contractual recurring revenue.

We could not find a rigorous, public study establishing 3:1 as the right threshold for physical-product ecommerce. It is repeated widely and sourced rarely. So reason from the math instead.

The ratio only means something if LTV is measured in contribution margin, not revenue. A brand with a 33% CM2 margin and a "3:1 LTV to CAC" calculated on revenue is exactly at break-even. Three dollars of revenue at 33% margin is one dollar of margin, which is what the customer cost.

The ratio is meaningless without a time window. A 3:1 ratio that takes three years to arrive will not help a business that runs out of cash in month eight.

A more useful statement of the rule for D2C:

• Margin-based LTV to CAC above 1.0 within your payback window means acquisition is profitable.
• Margin-based LTV to CAC of 1.5 to 2.0 at 12 months leaves room for fixed costs and error. That range is our reasoning from the math, not a published benchmark.
• A very high ratio usually means you are under-spending and leaving growth unclaimed.

How do you calculate LTV honestly for a young brand?

Use the cohort method. It needs no assumptions, only order data.

1. Group customers by the month of their first order. Each group is a cohort.
2. For each cohort, add up net revenue from all their orders at 30, 60, 90, and 180 days after first purchase.
3. Divide by the number of customers in the cohort. That is revenue LTV at each window.
4. Convert to contribution margin using your CM2 per order. That is margin LTV.
5. Compare margin LTV at each window against the blended CAC you paid in that cohort's month.

Shopify's customer cohort analysis report does steps 1 to 3 for you.

A worked example: one cohort over 180 days

Use the $60 skincare brand again. Suppose new-customer CAC rises to $40, close to the Triple Whale median. First orders carry the 10% code, so first-order CM2 is $26.63. Repeat orders are at full price, so repeat CM2 is $32.46 ($60.00 minus $15.00 product, $6.50 shipping, $2.50 pick and pack, $2.04 payment fee, and $1.50 returns allowance). The repeat rates are illustrative.

Days since first order	Cumulative orders per customer	Revenue LTV	Margin LTV	Margin LTV minus $40 CAC	Margin LTV to CAC
0	1.00	$54.00	$26.63	-$13.37	0.67
30	1.08	$58.80	$29.23	-$10.77	0.73
60	1.18	$64.80	$32.47	-$7.53	0.81
90	1.27	$70.20	$35.39	-$4.61	0.88
180	1.45	$81.00	$41.24	$1.24	1.03

At 180 days, revenue LTV to CAC is about 2.0 (81 divided by 40). That sounds acceptable. Margin LTV to CAC is 1.03. The cohort has only just paid back, six months after you spent the money. Payback lands between day 90 and day 180.

This is the picture behind "we are growing and still losing money." Customers do come back. But every new cohort ties up $13.37 per customer for most of six months, and the faster you grow, the more cash is tied up.

Why you should not project a three-year LTV

A brand with eight months of history has zero customers who have been around for three years. Any three-year LTV is an extrapolation, and the common formula (average order value x purchase frequency x customer lifespan) invents the lifespan.

Early cohorts also mislead. Your first customers are friends, early adopters, and people who were searching for exactly your product. They repeat more than the cold audiences you reach at scale. Later cohorts usually look worse.

Use the longest window you have actually observed. If you have six months of data, your LTV is your 180-day LTV. Extend it only when a cohort has lived through the longer window.

What repeat purchase rate should you expect?

It depends mostly on whether the product runs out. Consumables repeat. Durables mostly do not.

Public category data is thinner than most blog posts admit, and definitions vary (repeat within 90 days, within a year, or ever). The figures below are the ones we could trace to a named dataset.

Benchmark	Figure	Data and method	Source
Average repeat purchase rate, ecommerce brands	28.2%	Metrilo client data. Self-selected brands already using retention tooling, so it may read high	Metrilo
Cosmetics brands	25.9%	Same Metrilo research	Metrilo
Tea brands (lowest category in the study)	20.9%	Same Metrilo research	Metrilo
Subscribers vs one-time shoppers	Subscribers placed nearly 3 times as many orders	20,000 brands on Recharge	Recharge Subscription Trend Report 2026

The honest read: roughly one in four to one in three customers buys again at an average brand, and the spread by category is wide. If your plan needs 50% of customers to reorder within 90 days and you sell a non-consumable, the plan is wrong.

You will often see the claim that a 5% increase in retention raises profits by 25% to 95%. It traces to Frederick Reichheld of Bain & Company, as cited in Harvard Business Review. The research predates modern ecommerce and the range is very wide. Treat it as a reason to care about retention, not as a forecast for your store.

Compute your own number. In Shopify, the returning customer rate and cohort reports show it directly. Track 60-day and 90-day repeat rate by acquisition month, and by first product purchased. The first product often predicts repeat behavior better than the channel does.

How do you raise AOV without hurting conversion?

Average order value is the cheapest lever in the model because most costs per order are fixed. Shipping, pick and pack, the 30 cent payment fee, and CAC barely move when a customer adds a second item.

Run the worked example with a second $30 item added at a 25% product cost. Net revenue rises by $30. Product cost rises $7.50. Shipping might rise $1. The payment fee rises 87 cents. CM2 rises by about $20.63, from $26.63 to $47.26, on the same ad spend. CM3 goes from $4.63 to $25.26.

That is why bundles, thresholds, and post-purchase offers earn their place. The evidence on what works:

• Free shipping thresholds. In Sendcloud's 2025 consumer research, a survey of 8,000 consumers across eight European markets, 70.5% said they were likely to add items to reach a free delivery threshold. Set the threshold above your current AOV, not at it. Our guide to setting a Shopify free shipping threshold walks through the math.
• Shipping costs shown late lose the sale. Baymard Institute puts the average documented cart abandonment rate at 70.22% across 50 studies, and extra costs such as shipping, tax, and fees are the top reason shoppers cite once "just browsing" is set aside, named by roughly 4 in 10 US shoppers. A threshold turns that objection into a reason to add to the cart.
• Bundles and multi-packs. We did not find a rigorous public dataset on bundle uplift, so test it. The math favors any bundle where the second unit ships in the same box.

For context, the median AOV in Triple Whale's Meta dataset is $73.36. The conversion side of this equation is covered in our ecommerce conversion rate benchmarks guide.

One warning. Do not raise AOV with discounts that cost more than the margin they add. "Buy two, get 30% off" on a 49% CM2 product gives away most of the gain.

What should email and SMS contribute?

Owned channels are where payback gets shorter. An email costs almost nothing to send, so repeat revenue from email carries close to full CM2.

The published benchmarks agree on one point: automated messages do most of the work per send.

Benchmark	Figure	Data and method	Source
Share of email revenue from automations	30% of email revenue from 2% of sends	27 billion emails sent by 150,000 brands in 2025	Omnisend 2026 Ecommerce Marketing Report
Revenue per email, automated vs campaign	$2.87 vs $0.18	Same report	Omnisend
Placed order rate, flows vs campaigns	About 13 times higher in flows	Klaviyo customer base, 2026 benchmarks, more than 183,000 customers	Klaviyo
Click rate, flows vs campaigns	5.58% vs 1.69%	Same dataset	Klaviyo
Abandoned cart flow, revenue per recipient	$6.77 by email, $4.48 by text. Top 10% of email flows reach $13.70	More than 110,000 Klaviyo customers	Klaviyo

You will see "email should be 25% to 30% of revenue" quoted as a target. We could not trace it to a dataset. The share depends on how much you spend on ads, so it is a poor benchmark anyway. A brand that cuts ad spend will see its email share jump without email improving.

What to do with this: build the welcome, abandoned cart, browse abandonment, and post-purchase flows before you scale ads. They convert the traffic you are already paying for. Our walkthrough on Shopify abandoned cart recovery covers the setup. Then measure owned channels by what they do to 60-day and 90-day cohort LTV, not by attributed revenue share.

Do subscriptions fix unit economics?

They help when the product is truly consumed on a schedule. Recharge's 2026 report found subscribers placed nearly three times as many orders as one-time shoppers across 20,000 brands. More orders per customer means CAC is spread across more margin.

Two cautions. Brands in the same report raised first-order subscription discounts by 18%, so check that the discount does not push your first-order CM2 below zero. And subscribers churn. Model subscription LTV from your own cohort curves, the same way as above.

Why do profitable brands run out of cash?

Because profit and cash arrive at different times. You pay for inventory weeks or months before you sell it, and you pay for ads days before the order is delivered. A brand can have positive CM3 on every order and still be unable to pay its supplier.

The measure is the cash conversion cycle. Shopify's guide gives the standard formula.

Cash conversion cycle = days inventory outstanding + days sales outstanding - days payable outstanding

• Days inventory outstanding (DIO): average inventory divided by cost of goods sold, times 365. How long stock sits before it sells.
• Days sales outstanding (DSO): how long until you receive the money. For card payments this is a few days. For cash on delivery it can be weeks.
• Days payable outstanding (DPO): how long your supplier lets you wait before paying.

A worked example. You hold 90 days of stock, your processor pays out in 2 days, and your supplier wants payment in 30 days. The cycle is 90 + 2 - 30 = 62 days. Every dollar of product cost is locked up for two months.

Now grow. If monthly cost of goods rises from $20,000 to $40,000 and you keep 90 days of stock, inventory on hand goes from $60,000 to $120,000. You need $60,000 of new cash before you see a cent of the extra profit. Add a payback period on CAC and the gap widens.

The levers are few and practical. Order smaller quantities more often even at a slightly worse unit price. Negotiate payment terms before you negotiate price. Cut slow SKUs. Take pre-orders for launches. And slow ad spend when cash is tight, even if aMER looks good.

How much should you spend on ads at each stage?

Express the budget as a ratio of your own margin, not as a dollar figure someone else used. The three stages need different rules.

Stage	Goal	Budget rule	Stop or change when
Pre-launch	Know your numbers before buying traffic	No scaling spend. Build the order-level P&L. Calculate CM2, break-even ROAS, and break-even CAC	CM1 is under 50% or break-even ROAS is above 3. Fix price, cost, or AOV first
First 100 orders	Buy data, not profit	Set a fixed learning budget you can afford to lose. Size it from platform needs: about 50 conversions per ad set per week, times your target CPA	aMER is under half of break-even after 100 orders, or conversion rate is far below benchmark
Scaling	Grow contribution dollars, not revenue	Raise spend in steps while aMER stays at or above break-even, or while payback stays inside the window your cash can fund	aMER falls below break-even for two weeks running, or cash cover drops below your payback window

On the learning budget: Meta's guidance on the learning phase is that an ad set needs about 50 optimization events in a 7-day period to exit learning. If your target CPA is $30, one ad set optimizing for purchases needs roughly $1,500 a week to learn properly. Spread a smaller budget over five ad sets and none of them learn. With a small budget, run fewer ad sets, or optimize for a cheaper event such as add to cart until volume builds.

On scaling: watch marginal efficiency, not average. The next $1,000 of spend almost always buys customers at a worse CAC than the last $1,000. If doubling spend takes aMER from 2.6 to 2.2 and your break-even is 2.03, the average still looks fine, but the added spend came in near 1.8 (the second half must average 1.8 for the blend to be 2.2). You lost money on every customer the increase bought.

The target to optimize is total contribution dollars after ad spend. A lower MER with much higher volume can be the better business.

Paid is also not the only acquisition channel. Organic search and AI search bring customers with no media cost per click, which lowers blended CAC over time. Our D2C ecommerce SEO guide covers that side, and the D2C launch guide covers what to have in place before you spend at all.

India: where the math differs

Everything above applies in India. Three things change the numbers enough to need their own lines in your model. All figures in this section are India-specific.

Cash on delivery and RTO

Cash on delivery (COD) is still a large share of orders. In Fynd's Black Friday 2025 data, COD fell to 45.2% of orders from 49.9% a year earlier.

COD orders come back undelivered far more often. This is return to origin (RTO): the customer refuses the parcel or cannot be reached, and it ships back at your cost.

Benchmark (India)	COD orders	Prepaid orders	Source
Share of orders returned, FY25	Nearly 26%	Less than 2%	Shipway ShipNotes report, based on D2C shipments processed in FY2025, as covered by MediaInfoline
RTO rate	25% to 30%	2% to 3%	Razorpay

Both sources sell logistics or payments products, so read them as informed estimates. They agree closely with each other.

RTO hits unit economics three ways. You pay forward and reverse shipping on an order with no revenue. Your ad spend bought an order that never completed. And your stock is in transit for weeks instead of on the shelf.

A worked example: prepaid vs COD (in rupees)

Suppose a prepaid order earns ₹400 of CM2. A delivered COD order earns ₹360 after a ₹40 COD handling charge. An RTO costs ₹145 (₹70 forward shipping, ₹60 reverse shipping, ₹15 packaging). CAC is ₹300 per order placed. The inputs are illustrative. The RTO rate is the Shipway FY25 figure.

Line	Prepaid	COD at 26% RTO
CM2 on a delivered order	₹400.00	₹360.00
Share of orders delivered	100%	74%
Cost of each undelivered order		-₹145.00
Expected CM2 per order placed	₹400.00	₹228.70
CAC per order placed	₹300.00	₹300.00
Expected CM3 per order placed	₹100.00	-₹71.30
Effective CAC per delivered order	₹300.00	₹405.41

The same product, the same ad, the same CAC. The prepaid order makes ₹100. The COD order loses ₹71.30. A dashboard that counts orders placed will show both as wins.

In India, always compute ROAS and CAC on delivered revenue, not placed orders. Then work the levers: a small prepaid discount, COD confirmation by WhatsApp or phone before dispatch, a COD fee, and blocking COD for pin codes or customers with a history of refusals.

Payment gateway fees and GST

Razorpay's standard rate is a 2% platform fee, with 18% GST charged on the fee, which comes to about 2.36% of the transaction. International cards are 3% plus GST, about 3.54%. See the Razorpay pricing page for current rates by method.

Indian prices are shown inclusive of GST. Strip GST out of revenue before you calculate margin, or every percentage in your model will be overstated. ROAS computed on GST-inclusive revenue is not comparable with a margin computed on net revenue.

COD also stretches the cash cycle. Couriers remit COD collections days after delivery, so days sales outstanding is longer than for prepaid orders. Add that to the cash conversion cycle above.

The weekly unit economics dashboard

Ten numbers, checked every Monday, will tell you more than any attribution tool. Build them in a spreadsheet first.

Metric	Formula	Where to find the inputs	Reference point (source)
Net revenue	Gross sales minus discounts minus refunds, excluding tax	Shopify: Analytics, Finance summary	Your own trend
CM2 margin %	(Net revenue minus COGS, shipping, pick and pack, fees, returns allowance) / net revenue	Shopify cost per item, shipping and fulfillment invoices, payout reports	Above 45% makes paid acquisition workable (break-even table)
MER	Net revenue / total ad spend	Shopify revenue. Meta and Google billing totals	Must sit well above break-even ROAS
aMER	New-customer revenue / total ad spend	Shopify: first-time vs returning customer sales	At or above break-even ROAS (1 / CM2 margin %)
Blended new-customer CAC	Total ad spend / first-time customers	Shopify customer reports. Ad platform billing	Below first-order CM2, or inside your payback plan. Median Meta CPA is $38.99 (Triple Whale)
Platform ROAS	Platform-attributed revenue / platform spend	Meta Ads Manager, Google Ads	Median Meta ROAS is 1.88 (Triple Whale). Use for in-platform decisions only
AOV	Net revenue / orders	Shopify: Analytics	Median $73.36 in Triple Whale's Meta dataset
60-day and 90-day repeat rate	Customers with a second order within the window / customers in the cohort	Shopify: customer cohort analysis	Average repeat purchase rate 28.2% (Metrilo, no fixed window)
Return or RTO rate	Returned or undelivered orders / orders shipped	Shopify returns. Courier reports	US online returns 19.3% (NRF 2025). India COD RTO about 26% (Shipway FY25)
Cash cover	Cash on hand / (weekly ad spend + weekly inventory payments)	Bank balance. Purchase orders	Longer than your CAC payback period

GA4 is deliberately not the source for any of these. Use it for funnel questions. Use the store and the bank for money questions.

Seven mistakes that hide losses

1. Using product margin for break-even ROAS. The right input is CM2, after shipping, fees, and returns. The gap is often 20 points.
2. Trusting platform ROAS as profit. Platforms count returning customers and view-through conversions. In the worked example they claimed 89% of revenue between them.
3. Calculating LTV on revenue. A 3:1 revenue ratio at a 33% margin is break-even.
4. Projecting LTV beyond the data. Six months of history supports a 180-day LTV and nothing longer.
5. Ignoring discounts. A 10% code on a 49% CM2 product gives away about a fifth of the margin.
6. Counting placed orders instead of kept orders. Returns and RTO remove revenue but not the ad cost.
7. Scaling on average efficiency. The marginal dollar is what you are deciding on, and it is always worse than the average.

A 30-day profit audit

You can do this with a spreadsheet and exports from Shopify and your ad accounts.

Week 1: build the order-level P&L

• Export 90 days of orders. Calculate net revenue per order after discounts and refunds, excluding tax.
• Get the true landed cost per SKU, including freight and duties.
• Pull actual shipping, pick and pack, and packaging cost per order from invoices, not estimates.
• Calculate payment fees from payout reports.
• Calculate your return rate and cost per return. In India, calculate RTO rate separately for COD and prepaid.
• Result: CM1 and CM2 per order, overall and for your top 10 SKUs.

Week 2: find your break-even and your real CAC

• Calculate break-even ROAS (1 / CM2 margin %) and break-even CAC.
• Calculate MER, aMER, and blended new-customer CAC for each of the last 12 weeks.
• Compare aMER with break-even ROAS. Mark the weeks you acquired at a first-order loss.
• Add a one-question post-purchase survey to the order confirmation page.

Week 3: measure what customers are worth

• Build cohorts by first-order month. Calculate margin LTV at 30, 60, 90, and 180 days.
• Find your payback period. Compare it with how many weeks of ad spend your cash covers.
• Split repeat rate by first product purchased and by acquisition channel.
• Check that welcome, abandoned cart, and post-purchase flows are live and measured.

Week 4: act on the three biggest levers

• Fix the worst SKU: reprice it, bundle it, or stop advertising it.
• Set a free shipping threshold above current AOV and test one bundle.
• Cut or restructure campaigns whose new-customer CAC is above break-even with no payback evidence.
• Set a weekly budget rule tied to aMER and cash cover, and write it down.
• Put the ten dashboard metrics in one sheet and review it every Monday.

If the audit shows the site itself is losing the sale, with slow pages, a weak product page, or checkout friction, fix that before buying more traffic. A free website audit is a quick way to get a second set of eyes on it. And if you want the paid side or the store itself handled for you, our Shopify growth work is built around these same numbers.

Sources and methodology

Every figure in this guide comes from a named public source, listed below, accessed November 2026. Benchmarks from software vendors describe their own customer bases and are not random samples of all ecommerce brands. Worked examples use illustrative inputs and are labeled as such. Where a common rule of thumb has no rigorous public source, the text says so.

• Triple Whale, Facebook Ad Benchmarks by Industry. Median Meta ROAS, CPA, CPM, CTR, conversion rate, and AOV, with industry detail, August 2025 to July 2026. triplewhale.com
• Triple Whale, New Customer Cost per Acquisition documentation. Definition of new customer CPA. triplewhale.readme.io
• Triple Whale, Marketing Efficiency Ratio. MER definition and the inverted convention. triplewhale.com
• WordStream by LocaliQ, 2026 Google Ads Benchmarks. Average search CPC, US campaigns, April 2025 to March 2026, with 2025 edition comparison. wordstream.com
• WordStream by LocaliQ, Facebook Ads Benchmarks 2026. Traffic campaign CPC and click-through rate. wordstream.com
• SimplicityDX, press release, July 2022. Loss per new customer acquired, 2013 vs 2022. businesswire.com
• Paddle (ProfitWell), How is CAC changing over time. CAC increase over five years for subscription companies. paddle.com
• NRF and Happy Returns, 2025 Retail Returns Landscape. Total and online return rates, survey method. nrf.com
• Shopify pricing. Shopify Payments card rates and third-party transaction fees. shopify.com
• Stripe pricing. Standard card rate, international and conversion surcharges. stripe.com
• PayPal business fees. PayPal Checkout and standard card rates. paypal.com
• Razorpay pricing. Platform fee, GST on fee, international card rate. razorpay.com
• Razorpay, Cash on Delivery and the challenges. COD and prepaid RTO ranges in India. razorpay.com
• Shipway ShipNotes report (FY25), via MediaInfoline coverage. Share of COD and prepaid orders returned. mediainfoline.com
• Fynd Black Friday 2025 report, via BW Retail World. COD share of orders, 2024 vs 2025. bwretailworld.com
• Warby Parker, Form 10-K, FY2025. Gross margin and tariff commentary. sec.gov
• Hims & Hers Health, Form 10-K, FY2025. Gross margin. sec.gov
• Allbirds, Form 10-K, FY2025. Gross margin. sec.gov
• FIGS, fourth quarter and full year 2025 results. Gross margin. businesswire.com
• YETI Holdings, 2026 proxy statement. FY2025 adjusted gross margin. sec.gov
• David Skok, SaaS Metrics 2.0, For Entrepreneurs. Origin of the 3:1 LTV to CAC and 12-month payback guidelines. forentrepreneurs.com
• Harvard Business Review, The Value of Keeping the Right Customers (2014). Reichheld and Bain retention and profit claim. hbr.org
• Metrilo, Customer retention and DTC brands report. Average and category repeat purchase rates. medium.com
• Recharge, Subscription Trend Report 2026. Subscriber order frequency and first-order discount trend, 20,000 brands. getrecharge.com
• Omnisend, 2026 Ecommerce Marketing Report. Automation share of email revenue and revenue per email, 2025 data. omnisend.com
• Klaviyo, email marketing benchmarks. Flow and campaign click and placed order rates, 2026. klaviyo.com
• Klaviyo, abandoned cart benchmarks. Revenue per recipient for email and text. klaviyo.com
• Sendcloud, Why delivery costs are a dealbreaker (2025). Share of shoppers who add items to reach a free delivery threshold. sendcloud.com
• Baymard Institute, cart abandonment rate statistics. Average documented abandonment rate and reasons. baymard.com
• CNBC, February 2022. Meta's estimate of the 2022 revenue impact of Apple's privacy change. cnbc.com
• Adjust, ATT opt-in rates 2025. Industry opt-in rate, Q2 2025. adjust.com
• Meta for Developers, Ads Insights API metric availability updates (October 2025). Removal of 7-day view and 28-day view windows from January 12, 2026. developers.facebook.com
• Meta Business Help Center, About the learning phase. About 50 optimization events in 7 days. facebook.com
• Google Analytics Help, Advertising and attribution. Attribution models and settings in GA4. support.google.com
• Shopify Help Center, Customers reports. Customer cohort analysis and returning customer reports. help.shopify.com
• Shopify, Cash conversion cycle. Formula and component definitions. shopify.com