How to Start a Dropshipping Business: The Complete 2026 Blueprint.

Nobody wakes up wanting to buy from a dropshipping store. They see an ad, usually a short video on Meta or TikTok, for a product they did not know existed, and they buy on impulse at a price point between $25 and $70. That is the actual customer: a paid-social impulse buyer, not a search-intent shopper. This matters because it defines every other decision. Search-intent categories (replacement parts, printer ink, known brands) are owned by Amazon and by the brand itself, and you cannot outbid them on Google Shopping with a 30% gross margin. Your buyer has to be discovered, interrupted, and converted in one session, because they will not remember your store name tomorrow.

The second real buyer segment is the niche enthusiast who found you through organic content, a niche forum, a subreddit, or an email list. This customer converts at 2 to 4% instead of the 1 to 2% typical of cold paid traffic, returns less, and charges back less. Building that audience takes 6 to 18 months of content, which is why most people start with paid ads and then discover the ad math does not work. The stores that survive past year one almost always end up with a mixed traffic base: paid ads for volume, email and SMS for repeat purchases, and some organic content moat.

On size: U.S. e-commerce is well over a trillion dollars annually, but that number is irrelevant to you because Amazon, Walmart, and direct brands take the overwhelming majority. The addressable slice for an independent dropshipper is the long tail of products that are too new, too odd, or too seasonal for a large retailer to stock. There is no authoritative government dataset that isolates dropshipping revenue, so treat any national market-size figure you see as vendor marketing. What is verifiable: a small early store realistically does $0 to $5,000 in months 1 through 3, and $5,000 to $50,000 per month in months 4 through 12 if the marketing works. Many never clear the first band.

The competitive pressure in 2026 comes from two directions. Below you, Temu and Shein sell the same generic goods at prices you cannot match, so you cannot compete on a commodity item with a $9 landed cost. Above you, established DTC brands have better creative budgets and lower blended acquisition costs from repeat purchase. The viable position is in the middle: a product with a real hook, a differentiated offer or bundle, and enough perceived value to support a 3x to 4x markup over landed cost.

Dropshipping has no physical siting decision in the usual sense, but it has a harder one: which acquisition channel you build the business on, and that choice is as irreversible as signing a lease. Paid social (Meta and TikTok) gives you the fastest path to a first sale, often within days of launch, but you rent the audience and your unit economics move whenever auction prices move. Google Shopping and search capture demand that already exists, which converts better, but only works if people already search for your product category, which by definition rules out most novelty items. Marketplaces (Amazon, Walmart, eBay) give you traffic for free but take 8 to 15% in referral fees, hide the customer from you, and will suspend an account for late shipment metrics without much appeal.

The second siting decision is your legal home. Form the entity in the state where you actually operate, not in Delaware or Wyoming because a video told you to. If you live and work in California and form a Wyoming LLC, California will still consider you doing business in California, and you will owe the California registration plus the annual franchise tax anyway. You will now pay two states instead of one. The only people who benefit from out-of-state formation for a one-person online store are the formation services selling the package.

The third is where your supplier ships from. A supplier warehousing in the U.S. delivers in 3 to 7 days and cuts your refund rate materially. A supplier shipping from overseas delivers in 10 to 25 days and generates most of your customer service load. The overseas supplier often has better margins on paper. Run the numbers with a realistic refund rate attached to each and the U.S. option frequently wins, especially in categories where the buyer expects fast delivery. Many mature stores end the first year having moved their top two or three SKUs to a U.S. third-party warehouse and kept only the long tail on overseas fulfillment.

Finally, decide early whether you are building a single-product brand or a general store. General stores are cheaper to start and nearly impossible to build repeat purchase or email value into. Single-product and niche stores cost more in creative and take longer to validate, but they are the only version of this business that has a resale value at the end.

Start with a concrete order. A $49.99 product with a $15 landed cost from the supplier, including their shipping. Payment processing takes roughly 2.9% plus $0.30, so about $1.75. That leaves $33.24 in gross profit, a gross margin near 66%, which is the high end of the 20 to 50% band the category typically reports because many stores sell at lower markups. Now the deductions that never appear in the product research tool: your platform subscription and apps amortized per order, roughly $0.50 to $3.00 at low volume; your email and SMS software; and the sales tax you collect and remit, which is neutral only if you registered and are actually remitting.

Then the two costs that determine whether the business exists. First, customer acquisition. On cold paid social, expect a cost per acquisition between $25 and $60 for a $50 product in a competitive category. At $30, you keep about $3 on the first order. At $40, you lose $7. Second, refunds and disputes. Assume a 5 to 15% refund rate depending on category and delivery speed. Every refund costs you the product cost you already paid the supplier, the processing fee that is not returned, and often the return shipping. A single chargeback costs the order value plus a $15 to $25 dispute fee.

This is why the entire business hinges on average order value and repeat purchase, not on finding a better product. Raising average order value from $50 to $80 with a bundle, a two-pack, or a post-purchase upsell moves gross profit per order from $33 to roughly $50 while the acquisition cost stays close to flat, because you paid to acquire the same person. That single change turns a $30 acquisition cost from marginal into workable. Repeat purchase does the same thing on the second order, which costs you an email send rather than $30 of media.

Plan for negative contribution in months 1 through 3. Realistic early revenue is $0 to $5,000 per month, and creative testing burns $250 to $700 before you learn anything reliable. The $500 to $5,000 contingency reserve is not optional padding; it is what pays supplier invoices during the weeks when refunds and ad spend both land before the payment processor releases your funds, which is commonly a 2 to 7 day rolling delay and sometimes a 90-day reserve on a new high-risk merchant account.

At launch, the honest answer is one person: you. The owner-operator handles supplier communication, ad creative and media buying, order management, customer service, and the books. Cash wage cost is $0 and time burden is the highest it will ever be, roughly 25 to 50 hours per week for the first three months. The first thing to outsource is not customer service and not ads. It is bookkeeping, because a CPA or bookkeeper at $150 to $500 per month catches sales tax nexus problems and misclassified ad spend before they compound into a filing mess, and because founders reliably do this task last and worst.

The first real hire, usually somewhere between 100 and 300 orders per month, is a customer service representative, often part time and often a contractor. BLS reports a mean annual wage of $43,520 for Customer Service Representatives nationally, with a median of $39,680, a 10th percentile of $29,560, and a 90th percentile of $61,250. On a contract or part-time basis for e-commerce inbox and chat work, $15 to $25 per hour is the practical range. Expect roughly one support contact per 4 to 8 orders, higher if your supplier ships from overseas, so the hire arrives sooner than founders expect.

The second hire is an e-commerce operations assistant handling order routing, supplier follow-up, tracking exceptions, and returns, at roughly $18 to $30 per hour. If you ever move to a U.S. third-party warehouse or hold any inventory, note that BLS reported average hourly earnings of $32.63 for transportation and warehousing workers in July 2026, which is the realistic floor for fulfillment labor if you bring any of it in-house. Wage benchmarks for hand laborers and material movers and for marketing specialists were not verified in the primary OEWS tables for this edition, so treat any precise figure for those roles as uncertain.

The role founders hire too early is a freelance ad manager, typically $1,000 to $5,000 per month on retainer or a performance arrangement. Before you have a validated product with proven contribution margin, an outside media buyer is spending your budget to learn your business. Hire that role after you have a product that works and you are the constraint on scaling it, not before. And be clear about worker classification: a genuinely independent contractor sets their own hours and methods, and misclassifying a full-time customer service worker as a contractor creates back-tax and penalty exposure with the IRS and your state labor agency.

Days 1 through 14 are creative production, not spending. Produce 5 to 15 distinct ad assets: at least three different hooks in the first three seconds, a demonstration angle, a problem-agitate angle, and a social proof angle. Shoot the sample product yourself on a phone. Founder-shot demonstration video routinely beats supplier stock footage in the auction because it looks native rather than like an ad. If you launch with one creative you will spend $200 to $500 and learn only that it did not work, with no way to know why.

Days 15 through 30 are a structured test at $50 to $100 per day, roughly $250 to $700 total, on one platform only. Judge it on cost per click and cost per add-to-cart, not on sales, because you will not get enough purchase data at this spend to be statistically meaningful. You are looking for a single creative and audience combination that projects to a customer acquisition cost below your maximum allowable number. Simultaneously, capture email and SMS on every visitor with a first-order offer, because at this stage your list is the only asset you own outright.

Days 31 through 60 are fulfillment truth-telling. Personally track every order to delivery. Email each buyer and ask when it arrived and whether it matched the listing. This is unglamorous and it is the highest-value work in the first quarter, because it produces the delivery-time data that lets you write an accurate shipping policy, which is what keeps your chargeback rate under 1% and your payment processor from freezing your funds. Ask satisfied buyers for a photo review. Ten real photo reviews will lift conversion rate more than another $1,000 of media.

Days 61 through 90, do only one thing: raise average order value and set up flows. Add a bundle, a quantity break, and one post-purchase upsell. Build an abandoned checkout flow, a post-purchase flow, and a 30-day winback flow in software costing $0 to $150 per month. Then, and only then, increase paid spend, and increase it in steps of no more than 20 to 30% per week while watching contribution margin per order. Realistic outcome for a store that did all of this well: $5,000 to $15,000 in monthly revenue by month 4 to 6, at a thin or breakeven net margin. That is a normal, healthy result. The stores reporting $100,000 months in week six are selling you a course.