Starting an ecommerce business is the process of validating a product, registering the business, building a storefront, and acquiring the first customers, which helps a founder reach paying demand without the fixed cost of retail. The mechanical part takes weeks. What determines whether the business survives is what the founder does between the first sale and the first hire.
Look at the market before anything else. Ecommerce reached 17.1% of total US retail sales in Q2 2026, growing at roughly 1.8 times the rate of total retail.
TrueProfit puts roughly 60% of Shopify stores under $1,000 a month in revenue. The channel is growing, but most new entrants do not clear a wage.
The part that decides the outcome is cash, not revenue. A physical-product brand pays for inventory months before it collects the money from selling that inventory, and every hiring decision competes with the next purchase order for the same dollars. This guide covers what it costs, what margins are realistic, the seven steps to launch, and the operating model that carries a bootstrapped brand from the first sale to the first hire.
An ecommerce business is a company that sells products or services through an online store, which helps founders reach paying customers without a retail lease.
Two models matter for a bootstrapped physical-product brand. Business-to-consumer (B2C) is selling direct to the buyer, the core DTC model, with higher margins and a shorter cash cycle. Business-to-business (B2B) is wholesale and distribution: lower margins, larger orders, longer payment terms, and net-30 or net-60 waits on money you have already spent on inventory.
Which one you pick sets your gross margin and your cash cycle before you make a single marketing decision. A DTC skincare brand running 60% to 80% gross margin and a food and beverage brand running 35% or lower are not the same business, and they cannot carry the same team at the same revenue.
Yes, but the number every guide quotes is gross margin, and gross margin does not pay anybody. Net margin does. The gap between the two is where most of the real answer lives.
A sustainable gross margin for ecommerce runs 60% to 70%, with the most profitable stores around 60% to 65% (TrueProfit, 5,000-plus active Shopify stores, January 2025 to February 2026).
Median DTC net margin in 2026 is 3% to 10%, while ad spend alone consumes 20% to 35% of revenue.
Luca's worked example: a $75 apparel order nets $6.59, or 8.8% after subtracting cost of goods at 35%, ad spend at 25%, shipping and returns at 17%, platform fees at 2.9%, and payment processing at 3.3%.
Sourced from TrueProfit, aggregated from 5,000-plus active Shopify stores, January 2025 to February 2026, with business-model rows from its August 30 to September 5, 2026 weekly refresh, and Luca, 2026.
Dropshipping shows the highest gross margin and one of the lowest net margins in the table, because most stores running it pay for nearly every visit. Shopify's own guide to organic dropshipping shows the model can run on SEO, organic social, and email instead. That takes sustained content and list-building work. High gross margin without owned demand is an ad account with a storefront attached.
Source products and order samples before you commit to a full run. Sample orders run $30 to $100, and a first meaningful private-label order runs $3k to $10k-plus before packaging, shipping, and duties.
Check three things before you commit: a demand signal (existing search volume, an active community, rising category penetration), the unit-level margin math before ad spend, and the disqualifier test. Any product that cannot carry at least 60% gross margin at your realistic landed cost fails it, however much you like it. Then sell to 20 people in your target audience who you are not related to, before you design packaging. Samples first, a real audience second, packaging last.
Cash and margin decide the model, not preference. The four ecommerce business models below differ mostly in how much cash they demand up front, which is the first thing an ecommerce business plan has to settle.
Dropshipping needs no inventory. A supplier ships directly to the customer, and gross margin runs 65% to 70% but compresses to around 17% net because most dropshippers buy nearly all of their traffic. Print on demand needs zero upfront capital, with the same thinner unit economics shown there.
Private label carries the highest ceiling but needs $3k to $10k-plus of inventory cash committed before the first dollar of revenue. Wholesale and B2B run lower margins on larger orders with longer payment terms.
Pick the model whose cash requirement matches what you have. That requirement sets your working capital cycle for the life of the business.
Per the SBA, a sole proprietorship is a reasonable choice for a low-risk business and for an owner testing an idea, and the choice is not permanent. You can convert to an LLC as revenue and personal liability grow, though that conversion can carry state restrictions and tax consequences.
An LLC separates business and personal liability, protecting personal assets in most instances, and is the common structure once there is real inventory and real revenue at stake.
Registration runs roughly $130 to $1,000 depending on the state, so confirm the exact fee with your state's filing office before you file.
Three more things apply here: an EIN, which the IRS requires once you hire employees, form a partnership or corporation, or pay sales and excise taxes; a general business license, required in most US cities and counties even for a home-based operation; and sales tax nexus, triggered by physical presence or by crossing a state economic threshold. This is not legal or tax advice.
Most advice on how to sell online starts with the platform. Reverse that order. The channel decision comes first because it sets your fee structure. The channel decision comes first because it sets your fee structure. The full Amazon FBA fee stack runs 40% to 55% of the sale price once fulfillment, storage, inbound placement, advertising, and returns are counted, against the 15% referral fee most sellers budget for. That is defensible for a product with existing Amazon search demand and punishing for a brand trying to build owned demand.
Hosted e-commerce platforms start around $29 to $39 a month on Shopify, depending on the billing cycle.
A marketplace listing costs almost nothing upfront. A custom ecommerce website with real functionality runs $5k to $20k, money a pre-revenue brand should not spend.
Buy a theme instead of building one. Four things affect conversion at launch. Product photography: a phone and daylight beat nothing, and a professional shoot runs $20 to $75 per finished packshot and $100 to $500 or more per on-model image. Product copy that answers the objection instead of describing the item. A checkout with no added friction. Mobile, because that is where the traffic is.
Founders spend weeks on the homepage, but almost no paying customer sees it before the product pages.
Self-fulfill your first hundred orders. That is how you learn where the cost and the complaints come from, before you hand the job to someone else. Then move to a 3PL, the first outsourcing decision the brand makes and the template for every one after; pay for a function before you can justify a headcount for it.
Cover shipping supplies, and set a returns policy before the first return arrives. In Luca's worked example, shipping and returns absorb 17% of a $75 apparel order.
The founder does acquisition personally at this stage, and that is not a stopgap. Founder-led acquisition and support are how you learn what the customer values, which objections keep recurring, and where the messaging lands. That is market research no hire can reproduce.
Answer every support ticket yourself, write the first email marketing flows yourself, and read every review. New sellers commonly budget $300 to $1,500 a month on paid across Google Ads, Meta and TikTok. Some start at zero, building organic traffic through search engine optimization and social media marketing instead, but paid usually shortens time to first sale.
Ecommerce startup costs break into three honest bands, and the table below splits the leanest one further. A bare marketplace listing or print-on-demand launch with no inventory at all can start as low as $0 to $500, while a lean dropshipping store with its own platform and domain name runs $500 to $2,000.
Add real inventory to a branded store, and the range moves to $3k to $8k. A full brand-led launch, with a meaningful inventory buy, design, and paid media, runs $10k to $50k-plus.
Software is usually under 15% of true first-year spend, against the inventory, design, and paid-media costs below.
Inventory and acquisition absorb almost all of a founder's real first-year spend, so optimizing app subscriptions is optimizing the wrong number.
The cash conversion cycle binds harder than the launch budget. A physical-product brand pays for inventory 60 to 120-plus days before it collects the cash from selling that inventory, and median days inventory outstanding is about 129 days, roughly 2.8 inventory turns a year.
For example, a brand carrying $500k of inventory on a 90-day cycle has half a million dollars unavailable for three months.
Every hiring decision in the second half of this article is a cash decision before it is an org-design decision.
Sourced from 2026 industry cost breakdowns, presented as ranges rather than a single average given inconsistent secondary aggregators.
Four costs land outside every budget band above, and they land early. Sample orders run $30 to $100 before you commit to a production run. State registration runs $130 to $1,000. Product photography runs up to about $300 per product if you hire out. Business insurance and accounting software are small monthly lines that nobody forecasts and everybody pays.
A set of functions has to exist in some form for the business to operate at all; the list covers product sourcing and supplier management, brand and packaging, storefront, creative, paid acquisition, organic and SEO, email and SMS, support, fulfillment, inventory management and cash forecasting, bookkeeping, and compliance where the category demands it.
Who performs those functions, and through what engagement model, is the entire game. At launch, the honest answer for most is software or a freelancer, not a hire.
Sourced from Constant Hire, based on placement and search work.
Reach for engagement models in order; start with software, then a freelancer or offshore contractor, then an agency or fractional arrangement, then full-time. Force a real decision at each step instead of drifting into a salary.
Before the $250k threshold in the table above, only three things are worth real money: a freelance bookkeeper, a freelance designer or photographer who covers brand, packaging, and product images, and a 3PL. Everything else the founder does, or software does.
Regulated categories are the exception. Supplements, food and beverage, and cosmetics need episodic regulatory counsel and a contract manufacturer or testing lab from launch, since the FDA covers labeling claims and the FTC covers advertising claims, and both require prior substantiation.
A dedicated quality or regulatory hire can wait. Outside counsel handling label and claim review for a small product line runs $2k to $12k a year, and a substantiation file for a single claim runs $1.5k to $7.5k, cheap enough to outsource well past the first full-time generalist hire.
A role should open when a measurable bottleneck appears. Calendar timing and a competitor's org chart are poor substitutes for that signal. The diagnostic triggers are concrete key performance indicators (KPIs), support tickets per week, ad spend per month, orders per day, SKU count, channel count.
Ecommerce stores receive roughly 20 to 50 support tickets per 100 orders. Gorgias Ecom Lab data, measured at the $10M GMV band in March 2026, puts an electronics brand at about 46 per 100 orders against food and beverage at about 20. Gorgias notes nobody publishes a clean per-vertical table, so treat these as planning ratios rather than survey results.
Up to 30% of tickets are shipping-status questions that automation deflects. At 1,000 orders a month, the 20 to 50 per 100 rate gives 200 to 500 tickets, which one person with good macros handles. At 3,000 to 5,000 orders, one person does not.
Sourced from Constant Hire placement experience
The first ecommerce hire is a generalist ecommerce or operations manager, not a specialist and not a "head of." That hire becomes necessary when the founder can no longer run the site and daily operations alongside acquisition and product, usually between $250k and $1M in revenue, the band shown in the table above.
Brands defer retention too long.
Omnisend, which sells the software that runs these flows, found in its 2026 ecommerce report that automated flows produce 30% of email revenue from 2% of sends, earning 16 times more per send than scheduled campaigns.
Klaviyo, which sells the same category of software, reports the same shape across more than 183,000 customers, with flows at 41% of email revenue from 5.3% of sends.
Someone has to build those flows. In Constant Hire's placement work, brands with no dedicated owner sit at 10% to 15% of revenue from email and SMS, against 30% to 45% for brands with one. Vendor benchmarks land in the same range: Klaviyo puts top performers at 30% to 40%, and Darkroom's P90 set at 38% to 45%. That leaves 20 to 30 points of revenue share unclaimed every quarter.
A dedicated owner is not the same thing as a full-time hire. Constant Hire's position is that a full-time retention marketer makes sense past $20M in annual revenue, or past $10M with a subscription component. Below that, a freelancer, a Klaviyo specialist, or a fractional retention lead covers the function at a fraction of the loaded cost. The mistake is not hiring too late. It is leaving the channel unowned by anyone.
Brands put off hiring a bookkeeper just as often. Founders run on a shoebox of receipts long after clean books would help.
Three hires regularly land too senior, too early, per Constant Hire's own placement work. They are a "head of" title before there is a team to head, a senior brand or creative director hired at $2M when a performance executor was needed, and a full-time CFO before there are clean books to use.
The title-to-scope mismatch costs more than the salary does. A senior operator hired to build a function ends up executing it alone, leaves, and Constant Hire has seen the resulting rebuild cost the brand 6 to 12 months.
Colin Hale, Recruiting & Client Services Manager at Constant Hire, describes this from direct experience. "In our own search work, the roles that stall most often are growth-leadership searches at brands between $5M and $20M where the brief blends acquisition, retention, conversion, and merchandising into a single title. The search is hard because the role is three jobs.”
The nature of the work determines the engagement model. Episodic, low-context work (a theme build, a photoshoot, a logo) belongs with a freelancer or an agency, while continuous, high-context, cross-functional, customer-facing work belongs in-house, spanning P&L, support, and inventory.
A freelancer executes a scoped task. A fractional executive supplies senior judgment and ownership a few days a month on an open retainer, typically 8 to 40 hours.
A fractional engagement costs a monthly retainer and nothing else. A full-time employee costs considerably more than base salary.
A full-time US employee costs roughly 25% to 35% over base salary once payroll taxes, benefits, and software are counted, based on Constant Hire placement data. The long-standing MIT benchmark from senior lecturer Joseph G. Hadzima Jr. puts the same figure at 1.25 to 1.4 times base salary.
Recruiting cost is the part founders leave out entirely. Filling a specialist ecommerce role carries internal time, vacancy loss, and mis-hire risk well beyond the job-board invoice, and an open seat runs roughly $500 a day. That multiplier does not apply to contractors, who cost their invoiced rate, which is why early-stage brands lean on them and why the table below does the math this way.
Sourced from Constant Hire, using Hadzima's loaded-cost math.
A boutique agency at $8k a month costs $96k a year. A full-time performance marketer on a $110k base costs roughly $143k loaded.
The switch happens in two steps, not one. Under $25,000 a month in spend, outsource the whole function and keep the cash in inventory. Between $25,000 and $100,000, run a hybrid: bring creative in-house, because creative fatigue is the bottleneck at that volume, and keep the agency on media buying and tracking. Past $100,000 a month, percentage-of-spend fees at 10% to 20% start exceeding the loaded salary and bringing media in-house pays.
A founder who keeps a "freelancer" on continuous, directed, full-time work on core brand activity is creating a misclassification liability, and the contract label does not protect them.
Three tests can apply. They are the IRS common-law test, the Department of Labor's economic-reality test, which a February 2026 proposed rule would rebuild around control and opportunity for profit or loss, and a state ABC test, strictest in California. The Department no longer applies the 2024 rule in its own investigations, but the 2024 rule still governs private litigation, so the underlying exposure has not gone away.
Move a contractor to W-2 once the facts of the relationship point that way, and treat continuous full-time work on core activity as one of those facts. This is not legal advice, and the federal posture has been shifting.
Below $1M, the team is the founder plus one or two contractors. Crossing $1M moves a brand to three to five full-time people plus one or two contractors, covering conversion, acquisition, and operations. At $5M to $15M it is 8 to 15 with channel-specific leads, and past $15M it is 15 to 30-plus across departments (Constant Hire, 2026).
The lag signal at $1M to $5M is not headcount. It is contribution margin compressing while revenue grows, which means nobody owns on-site conversion or operational cost control. Freeze hiring and audit for work a contractor or automation could absorb before adding a salary against it. Org shape, reporting lines, and function mapping past $3M live in ecommerce team structure.
The hiring question underneath every threshold in this article is the same one: can contribution margin after ad spend and fulfillment carry a fixed salary without starving the next inventory buy? If the answer is no, the next move is a contractor or a fractional engagement, not a hire. A software company that adds a salary spends that salary. A physical-product company spends it on top of the cash already locked inside inventory.
Constant Hire places pre-vetted ecommerce and DTC operators, including paid media leads, lifecycle marketers, and ecommerce managers, with first interviews inside five business days. When the trigger for your first real hire fires, that is the search we run.
A bare marketplace listing or print-on-demand launch runs $0 to $500. A lean dropshipping store runs $500 to $2,000. Stocking a branded store puts you at $3k to $8k, while a full brand-led launch with paid media reaches $10k to $50k-plus. Inventory and acquisition absorb most of a founder's real first-year spend.
Not to start. Per the SBA, a sole proprietorship suits a low-risk business and an owner testing an idea, and the choice is not permanent. An LLC separates personal and business liability and is worth forming once you hold real inventory or real revenue. Registration runs roughly $130 to $1,000 by state. This is not legal advice.
A generalist ecommerce or operations manager fills that role, not a specialist or a "head of" title. The trigger is the founder no longer being able to run the site and daily operations alongside acquisition and product, usually between $250k and $1M in revenue. Before that, a 3PL, a freelance bookkeeper, and a freelance designer cover more ground per dollar.
Longer than the revenue suggests. A physical-product brand pays for inventory 60 to 120-plus days before it collects the cash from selling it, and median days inventory outstanding sits near 129 days. That means the cash from your first inventory buy typically returns in month four or five, not month one. Roughly 60% of Shopify stores earn under $1,000 a month, so a founder should plan to fund the first two inventory cycles without taking a wage.
Run the break-even before you choose. Under roughly $75,000 a month in ad spend, an agency or freelancer wins on cash and carries no ramp or severance risk. Above $75,000 to $100,000 across multiple channels, a percentage-of-spend arrangement starts exceeding a loaded in-house salary and bringing media in-house begins to pay. For creative, the flip sits near 25 to 35 assets a month. Below those thresholds, the freelancer is not a compromise. It is the correct answer.
Yes, with one constraint that decides whether it works. Fulfillment goes to a 3PL, support runs on a helpdesk with good macros, and email runs on app automation. What cannot be delegated cheaply is founder-led acquisition and customer contact, because that is how you learn which objections recur and where the messaging lands. If you cannot personally answer support tickets and read every review in the first hundred orders, the validation loop does not close and no amount of software substitutes for it.
Top talent on your calendar in under 5 days.
