An ecommerce marketing agency that counts profit, not ROAS
Direct answer: Ecommerce marketing that works manages first-order profit, not platform ROAS: contribution margin after cost of goods, shipping, returns and payment fees decides what you can pay for a customer. We benchmark cart abandonment, return rates and repeat purchase against 2026 data and build the product page, checkout and email flows that turn one order into a second.
Platform ROAS is the most flattering number in ecommerce and the least useful. We work first-order contribution, returns and repeat purchase, because that is what decides whether growth makes you money or just makes you busy.
A thousand carts, and what is left at the end
Cart abandonment has hovered near 70% since 2006. It has barely moved through two decades of payment innovation, which tells you the problem is not technology. It is what the checkout asks people to accept.
The checkout autopsy
Start with 1,000 people who added something to a cart. Follow what survives.
Why the 702 left
Baymard's US survey of 4,329 adults, excluding the 42% who were simply browsing. Tap any row for the fix. Percentages total more than 100 because people could pick several reasons.
Baymard Institute, US survey of 4,329 adults on reasons for cart abandonment, excluding the 42% who reported they were only browsing. Baymard also estimates an average conversion uplift of about 35% is available from checkout design changes alone, which is roughly $260 billion in recoverable orders across the US and EU. Device split from Dynamic Yield's benchmark network.
Seven of those ten reasons are choices someone made in your checkout. That is the argument for treating checkout as a conversion optimization project rather than a development ticket, and it is why we start here rather than in the ad account. Traffic bought into a checkout leaking at these rates is the most expensive mistake in ecommerce.
What did you actually make on the first order?
Revenue is not profit and ROAS is not either. Set your real numbers and this returns your contribution per order, your first-order profit or loss, your true break-even ROAS, and how many orders one customer has to place before you are ahead.
Median paid-traffic AOV in DTC ran near $74 in 2025.
Landed cost of goods including duties.
Pick, pack, box and outbound postage, net of anything the customer pays.
Ecommerce averages 19 to 20%. Apparel runs 20 to 40%, electronics 10 to 11%, beauty around 8%.
Total ad spend divided by new customers. Category averages run near $53 in food and drink and $120 or more in luxury fashion.
Returns are modeled as the refunded order plus outbound shipping, reverse logistics and restocking, using a $20 handling assumption on top of your shipping figure. Published processing costs run $10 to $65 per return. Break-even ROAS is one divided by contribution margin, which is arithmetic rather than opinion.
Break-even ROAS is a margin question
The ROAS you need has nothing to do with your industry and everything to do with your contribution margin. The dot is you.
Where returns take the margin
Returns do not cost you their percentage. They cost you their percentage plus the shipping both ways and the handling, which is why a 25% return rate can remove around 70% of unit contribution margin.
Returns are the quietest way to lose money
Every brand knows its return rate. Almost none of them know what a return actually costs, because the loss is spread across shipping, labor, restocking and inventory that comes back unsellable.
Average ecommerce return rates now run 19 to 20%, roughly two to three times the 5 to 9% that physical retail sees. The spread by category is enormous: apparel sits near 25% and can reach 40%, footwear around 18%, home and furniture 15 to 20%, electronics 10 to 11%, and beauty closer to 8%. Comparing your rate to the blended average is close to meaningless.
Each return costs somewhere between $10 and $65 to process before you count the outbound shipping you already paid. That is why the margin effect is so much larger than the headline percentage suggests, and why a 25% return rate can remove around 70% of unit contribution margin rather than 25% of it.
The market has noticed. In 2026 roughly 72% of US retailers charge some kind of return fee, up from 41% in 2023, and 53% of those report their return rates falling as a result. That is a real lever, and it is also a conversion risk, because an unsatisfactory returns policy is one of the ten reasons people abandon carts in the first place. Whether it is worth it comes down to your contribution margin, which is exactly the number the calculator above gives you.
The practical work is narrower than most people expect: better size and fit information, photography that shows scale and true color, product detail copy that pre-empts the reason for the most common return, and post-purchase messaging that sets delivery expectations. None of it is glamorous. All of it shows up in the margin line.
The second order decides whether the first one was worth buying
If your first order barely breaks even, and for most brands it does not, then the entire business rests on customers coming back. The data on when they come back is unusually clear, and most brands are asleep during the window.
When the second order actually lands
Across 40,397 repeat buyers, half came back inside 30 days and three quarters inside 90. The shaded strip is where most post-purchase email flows go silent.
Read that alongside the repeat rate itself. Across roughly 156,000 DTC customers the average brand sees about 18.8% of first-time buyers place a second order. By category, consumables run 25 to 45%, supplements near 38%, apparel 12 to 32%, home and durable goods 7 to 18%, and luxury fashion under 10%.
Two conclusions follow, and they point in opposite directions depending on what you sell. If you sell consumables, retention is the business and your acquisition budget should be sized against a customer who buys four times a year. If you sell durables or luxury, the second order is a bonus, and your ceiling on acquisition spend is set by what a single order earns, which puts first-order contribution margin at the center of everything.
The tactical gap is the same in both cases. Median time to a second purchase clusters around 15 to 35 days, while most post-purchase email sequences finish at day seven to fourteen. Brands go quiet in the exact window when the second order is being decided. Fixing that is email marketing and marketing automation work, and it is usually the cheapest incremental revenue available.
Why the ROAS in your ad account is not the ROAS in your bank
Every ad platform reports the revenue it can claim credit for. None of them subtract your product cost, your shipping, your payment fees or your returns, and none of them know what the other platforms also claimed.
Two things go wrong at once. Platform-reported return on ad spend has been measured overstating true return by roughly 2.3 times, because multiple channels each claim the same conversion. And the ROAS number itself says nothing about profit, because break-even ROAS is simply one divided by your contribution margin. A brand at 20% contribution margin needs 5.0 times just to stand still. A brand at 50% breaks even at 2.0. The same 3.0x is a triumph for one and a slow bleed for the other.
That is why the phrase good ROAS is close to meaningless without your margin next to it, and why we report blended marketing efficiency and contribution after ad spend instead. Profitable brands typically run overall marketing efficiency somewhere between 2.5 and 4 times, but the honest version of that sentence is that it depends entirely on the margin structure underneath.
The wider context makes this urgent rather than academic. Acquisition costs have risen sharply over the last several years, with one analysis of more than 80 ecommerce clients putting the increase at 222% across eight years, while median DTC contribution margin fell from about 35% in 2021 to about 22% in 2025. Rising costs met falling margins. Brands that only watch platform ROAS have been walking into that squeeze with a metric that cannot see it.
What an ecommerce marketing engagement covers
In roughly this order, because spending into a leaking checkout is how brands turn a marketing budget into a rounding error.
Checkout and product page
Field reduction, price transparency, guest checkout, wallet payments, mobile-first testing on real mid-range devices. Baymard puts the available uplift from checkout design alone at about 35%. This is conversion optimization applied where the money changes hands.
Paid acquisition against contribution
Shopping, Performance Max, Meta and search managed to your real break-even ROAS rather than a platform number. See PPC management and Google Ads management.
Abandoned cart and post-purchase flows
Recovery sequences for the 70% who leave, then a post-purchase program that stays awake through the 15 to 35 day window when the second order happens. Covered under email marketing.
Retargeting that respects intent
Recovery works far better on high-intent traffic than on cold social audiences, so we segment by source rather than blasting everyone who bounced. See retargeting.
Creative that carries the product
Product video, unboxing, demonstration and scale-accurate photography direction. Better product presentation lifts conversion and cuts returns at the same time, which is rare. See video production.
Returns and margin work
Fit guidance, expectation-setting copy, policy structure and the fee question. Not marketing in the traditional sense, and often the largest single lever on profit per order.
Search visibility and AI-search citation for product and category pages run through Eye To Ad Media, our search arm. If your store needs custom tooling that a theme cannot handle, such as a configurator, a subscription portal or a wholesale ordering system, that is web app development, built in-house.
The through line across all of it is the same principle we apply as a conversion rate optimization agency: improve what happens to the traffic you already pay for before buying more of it, because every point of conversion improvement makes every future ad dollar worth more.
When we are the right call, and when we are not
Talk to us if
- You are growing revenue and cannot work out why the bank balance is not growing with it
- Your checkout has more than eight fields or hides shipping until the final step
- Returns are above your category average and nobody owns the number
- Your post-purchase email stops before day fourteen
- You are scaling paid spend against a platform ROAS target nobody has checked against margin
- You sell consumables and your repeat rate is under 25%
Do not hire us if
- Your contribution margin is negative before acquisition cost, which is a pricing or sourcing problem first
- You want us to promise a ROAS number before seeing your margin structure
- You cannot give us order-level data including returns
- You are pre-launch with no sales history to work from
- You want SEO only, which belongs with a specialist search team
- Nobody on your side can ship a checkout change this quarter
Why us
Search Converts has been building and running commercial websites since 2012, and we maintain a working portfolio of several hundred properties ourselves. We are not describing checkout friction from a case study. We have built checkouts, watched people abandon them, and rebuilt them.
The founder also operates a home-services company outside the agency, which matters here for an unglamorous reason: it means the person reviewing your numbers has had to price a job knowing what the materials, the delivery and the callbacks actually cost. Contribution margin is not an abstraction to someone who has run a business where getting it wrong shows up in the bank account the same month.
We sell every marketing channel, so there is no incentive to make any one of them your answer. When the honest recommendation is to stop scaling spend and fix the checkout, or to raise prices, or to charge for returns, we say it. That advice costs us media revenue, which is exactly why it is worth listening to.
What you will not find on this page is a wall of brand logos or invented case study numbers. Every figure quoted here is published research with the source named. The first thing you get from working with us is the same analysis applied to your own numbers instead of somebody else's.
Ecommerce marketing questions, answered with numbers
An ecommerce marketing agency grows online store revenue across acquisition, conversion and retention. In practice the work splits into three jobs: buying traffic that can pay back at your margin, fixing the checkout and product pages so more of that traffic converts, and building the post-purchase program that produces the second order. Search Converts works all three against contribution margin rather than platform-reported return on ad spend, because a store can scale revenue and lose money at the same time without either party noticing for months.
There is no universal answer, because break-even return on ad spend is simply one divided by your contribution margin. A brand at 20% contribution margin needs 5.0 times just to break even, while a brand at 50% breaks even at 2.0 times. The same 3.0x result is profitable for one and loss-making for the other. Platform-reported ROAS has also been measured overstating true return by roughly 2.3 times because several channels claim the same conversion, so the number in your ad account is not the number in your bank. Calculate your own break-even first, then judge campaigns against it.
70.22%, based on Baymard Institute's meta-analysis of 50 separate studies, and the figure has been stable near 70% since around 2006. Mobile runs meaningfully worse than desktop, at roughly 80% against 69% across Dynamic Yield's benchmark network. Baymard measures shoppers who added at least one item to a cart rather than all browsing sessions, which is why it is a more useful benchmark than site-wide conversion rate.
Excluding the 42% who were only browsing, Baymard's US survey of 4,329 adults ranks the reasons as: extra costs too high at 39%, forced account creation at 19%, slow delivery at 18%, not trusting the site with card details at 18%, a checkout that was too long or complicated at 17%, and not being able to see the total cost upfront at 16%. Seven of the top ten are choices made in your own checkout. Baymard estimates an average conversion uplift of about 35% is available from checkout design changes alone, which is roughly $260 billion in recoverable orders across the US and EU.
Baymard's large-scale checkout usability testing puts an ideal flow at 12 to 14 form elements, which is roughly 7 to 8 actual input fields once labels and other elements are counted. Most stores run considerably above that. Enabling native mobile payment such as Apple Pay and Google Pay is reported to cut mobile abandonment by 8 to 12 percentage points on its own, largely because it removes the typing entirely on the device where abandonment is worst.
Around 19 to 20% overall in 2026, roughly two to three times the 5 to 9% rate seen in physical retail. Category variation is large: apparel runs near 25% and can reach 40%, footwear about 18%, home and furniture 15 to 20%, electronics 10 to 11%, and beauty closer to 8%. Each return costs between $10 and $65 to process before outbound shipping is counted, which is why a 25% return rate can remove around 70% of unit contribution margin rather than 25%.
Roughly 72% of US retailers charged some form of return fee in 2026, up from 41% in 2023, and 53% of those reported reduced return rates as a result. It works. It also has a cost, because an unsatisfactory returns policy is one of the ten reasons people abandon carts, cited by about 12% of abandoners. Whether the trade is worth it depends on your contribution margin and your category. Run the numbers before copying a competitor, because a brand at 45% margin and a brand at 18% should reach opposite conclusions.
The average DTC brand sees about 18.8% of first-time buyers place a second order, based on analysis of roughly 156,000 DTC customers. Category matters enormously: consumables should target 25 to 45%, supplements run near 38%, apparel 12 to 32%, home and durable goods 7 to 18%, and luxury fashion under 10%. Comparing a supplements brand to a furniture brand on this metric compares two different businesses. Track it by monthly acquisition cohort rather than in aggregate, or improvements and seasonal mix will be indistinguishable.
Send us five numbers and we will tell you the truth
Average order value, product cost, shipping cost, return rate and what you pay for a customer. That is enough for us to say whether your growth is profitable, where the leak is, and what we would fix first. No obligation, and no pitch deck.
Your information is never sold or shared. We respond within one business day.
Related pages
The same principles, applied to neighbouring industries and the channels that feed them:
- SaaS marketing — SaaS marketing agency that works the unit economics, not the lead count.
- B2B marketing — B2B marketing agency that plans for the 95% of buyers not in market yet.
- Google Ads management — Google Ads management from a PPC agency that fixes account structure before bidding.