Picking a second country is the most expensive decision a small seller makes with the least evidence. Domestic growth has a feedback loop you can read weekly. A new market gives you nothing for months, then hands you a customs bill, a returns backlog and a payment method you have never heard of, all in the same week. The sellers who survive that first expansion are rarely the ones with the biggest budget. They are the ones who chose a market that matched what their business already did well.
In short
- Demand is the cheapest thing to find and the least useful on its own. Almost every market has buyers for almost every product. What separates a good first market from a bad one is the cost of serving it, not the size of the addressable audience.
- Landed cost, not list price, decides whether you have a business. Duty, VAT or sales tax, freight, last-mile delivery, payment fees, currency spread and returns can add 25 to 60 percent to a product’s delivered cost depending on category and destination.
- Rules changed materially in 2025 and 2026 and are still moving. The US ended duty-free treatment for most low-value parcels and the EU has been reworking its own low-value import regime, so any duty threshold you memorized before 2025 is worth re-checking at the official source.
- Score three to five candidate markets, then commit to one. A scored shortlist beats intuition, and running two new markets at once usually means running both badly.
- Your operating model is a selection criterion. If you cannot support a language, hold local stock or accept the dominant local payment method, that market is harder for you than the market data suggests.
This guide walks through how to choose international market candidates with a repeatable method: what to measure, what to ignore, how to price the true cost of entry, and where first-time exporters most often get it wrong. It sits inside our broader work on selling on global e-commerce marketplaces, which covers the channel side of the same question.
Why market selection decides your first two years abroad
Expansion failures are usually attributed to demand. In practice they are attributed there because demand is the only variable most teams measured. The costs that actually sink a first international market appear after launch: a returns rate that doubles because sizing conventions differ, a customs classification dispute that freezes inventory, a payment method with a 4 percent fee that your pricing never accounted for.
The asymmetry matters because a first market is not a test. It is a commitment of working capital, catalog work, support hours and, most importantly, management attention. A small team can hold roughly one new market in its head at a time. Spend that attention on a market where your existing advantages transfer, and the learning compounds. Spend it on a market chosen from a population chart, and you spend a year discovering things a two-week scoring exercise would have told you.
The three questions that actually rank markets
Strip the exercise down and three questions do most of the ranking work. Can buyers there find and want what you sell? Can you deliver it at a price that still leaves margin after every cost of crossing a border? Can you support the customer afterward without building a new company?
Most first-time exporters answer the first question rigorously and the other two by assumption. Reversing that order is the single highest-return change you can make to the process. Demand research is cheap and abundant. Cost-to-serve and cost-to-support research is tedious, specific and where the decision actually lives.
Why “biggest market” is usually the wrong default
Germany, Japan and the United States show up at the top of nearly every market-size ranking, which is exactly why they are crowded. A large market with entrenched local players, high customer service expectations and strict consumer protection rules can be a harder first step than a mid-sized neighbor where your shipping lane is short and your brand has no incumbent to displace.
Size tells you the ceiling. It says nothing about the slope. For a first expansion, the slope matters more, because the point of market one is to build a repeatable playbook you can apply to markets two and three.
Key terms every first-time exporter needs
The vocabulary here is not decoration. Most costly mistakes trace back to a term the seller thought they understood.
Landed cost is the total cost of getting one unit into a customer’s hands in the destination country: product cost, outbound freight, insurance, duty, import taxes, customs brokerage, last-mile delivery and any handling fee the carrier adds. If your pricing model stops at product cost plus shipping, it is not a landed cost model.
HS code is the Harmonized System classification that determines the duty rate applied to your product. It is a globally standardized structure at the six-digit level, with countries adding further digits. Misclassification is one of the most common and most expensive errors in small-seller importing, because it can trigger back-duty and penalties long after the sale.
Incoterms are the standardized delivery terms published by the International Chamber of Commerce that define who pays for freight, who bears risk at each stage, and who is the importer of record. The two that matter most to direct-to-consumer sellers are DDP, where you deliver duty paid and the customer sees no surprise bill, and DAP, where the customer is billed for duty and tax on delivery. Choosing DAP to protect your margin is a reliable way to generate refused deliveries.
De minimis refers to a value threshold below which a country waives duty, and sometimes tax, on imported parcels. This concept changed dramatically for US-bound shipments: the long-standing $800 exemption under Section 321 was suspended and then repealed for most origins, a change the US Court of International Trade addressed in litigation we covered when the de minimis repeal survived its first major court test. Any figure you carry from memory here should be verified against current US Customs and Border Protection guidance before you price a single order.
Importer of record is the legal entity responsible to the destination customs authority for correct declaration and payment. If that entity is you, you have compliance obligations in that country. If it is your customer, they have them, and they will not enjoy the discovery.
Distance selling and VAT registration thresholds determine when you must register for and collect local consumption tax rather than relying on an import-level mechanism. The EU’s Import One-Stop Shop and the UK’s post-Brexit rules both changed how low-value goods are taxed, and both are administered by named authorities whose published guidance is the only reliable current source.
How to score candidate markets in practice
The method below takes two to three weeks of part-time work and replaces roughly a year of expensive learning. Start with a long list of eight to twelve countries, cut to three to five on hard filters, then score the survivors.
Step one: apply elimination filters first
Filters are binary and fast. Run them before any scoring, because they remove the majority of the list at near-zero cost. Eliminate any market where your category faces an import prohibition or a licensing regime you cannot satisfy, where you cannot legally sell without a local entity or representative, where no carrier offers a delivery service you can afford, or where the language burden exceeds what you can staff.
Category-specific rules catch people out. Cosmetics, supplements, electronics with radio components, children’s products and anything battery-powered all carry destination-specific requirements. A market that requires product registration before first sale is not disqualified forever, but it is disqualified as a first market for a small team.
Step two: score the survivors on weighted criteria
Give each surviving market a 1 to 5 score across weighted criteria, and be honest about the weights. The weighting below reflects what actually predicts first-market outcomes for small sellers, but adjust it to your category rather than adopting it blindly.
| Criterion | Weight | What a 5 looks like | What a 1 looks like |
|---|---|---|---|
| Demand evidence for your specific SKUs | 15% | Existing unsolicited orders, search volume for your exact product type, comparable sellers visible | Category demand inferred only from population and GDP |
| Landed cost as a share of retail price | 25% | Under 20% added cost, duty rate low or zero for your HS code | Over 45% added cost, high duty plus high last-mile |
| Logistics maturity and transit time | 15% | 3–7 day tracked delivery, multiple carrier options, returns lane exists | 14+ day transit, single carrier, no viable returns path |
| Payment method fit | 10% | Cards or a method you already accept dominate | Local wallet or bank transfer dominates and you cannot accept it |
| Regulatory and tax burden | 15% | Clear published guidance, no local entity needed, simple registration | Local representative required, product registration, opaque rules |
| Language and support load | 10% | You already operate in the language, or English is standard for your buyer | Full localization plus native-language support required |
| Competitive whitespace | 10% | No entrenched local equivalent, marketplace listings thin | Three strong local incumbents with better logistics |
Score the shortlist, then look at the spread. If two markets finish within a few points of each other, the tiebreaker should be operational: which one can you serve with the fewest new systems? A market that runs on your existing stack, carrier account and support language will generate a usable playbook faster, and the playbook is the actual deliverable of market one.
Step three: pressure-test the top choice before committing
Before you commit inventory, run a bounded test. Ship 30 to 50 real orders into the market at your intended price and Incoterm. Measure delivery time variance, customs holds, refused deliveries, support ticket volume per order and the return rate. That test costs a few thousand dollars and answers questions no dataset will.
Pay particular attention to variance rather than averages. An eight-day average transit time with a two-day standard deviation is a business. The same average with occasional three-week outliers is a support problem that will consume your margin in refunds and chargebacks.
What landed cost actually includes
Most first-time exporters build a landed cost model with three lines: product, freight, duty. The real model has closer to ten, and the missing seven are where margin disappears.
| Cost line | Typically missed? | Notes for first-time exporters |
|---|---|---|
| Product and packaging | No | Destination packaging or labeling rules may add cost per unit |
| Outbound freight | No | Volumetric weight often prices the parcel, not actual weight |
| Duty | No | Depends on HS code and origin; verify the rate with the destination customs authority |
| Import VAT, GST or sales tax | Often | Frequently confused with duty; different base, different remittance mechanism |
| Customs brokerage and clearance fees | Usually | Charged per shipment; brutal on low-value single parcels |
| Carrier disbursement or advancement fee | Almost always | Carriers charge a fee for fronting duty and tax, often a flat amount plus a percentage |
| Payment processing and cross-border card fees | Usually | Cross-border and currency conversion surcharges stack on the base rate |
| FX spread | Almost always | The gap between the mid-market rate and the rate you actually receive |
| Returns and reverse logistics | Almost always | International returns can cost more than the original outbound leg |
| Support cost per order | Almost always | New markets generate materially more tickets per order in the first six months |
The returns line is the one that kills apparel
Return rates in fashion and footwear run high in every market, and cross-border returns amplify every part of that cost. You pay to bring the item back, or you pay a local partner to receive and reprocess it, or you write it off entirely. Sellers who model returns at their domestic rate and their domestic cost consistently overestimate first-year margin in apparel by a wide margin.
There is a structural answer, and it is the reason many cross-border sellers eventually stop shipping direct. Once volume justifies it, holding local stock converts an expensive international parcel into a cheap domestic one and makes returns tractable. We looked at that shift in detail when examining why direct parcel models are giving way to domestic fulfillment, and the economics that drive it apply well below the scale of the companies in that story.
Pricing decisions that follow from the model
Once the full model exists, three pricing decisions become mechanical rather than intuitive. Whether to ship DDP and absorb duty into the price, whether to set a free-shipping threshold that pushes average order value above the point where the fixed clearance costs stop hurting, and whether some SKUs simply should not be offered in that market at all.
That last one is underused. A catalog that works domestically does not have to cross the border intact. Cutting the heavy, low-margin and high-return SKUs from your international assortment often turns a marginal market into a viable one.
Common mistakes and how to avoid them
Choosing by market size instead of by fit
The largest market is the one every competitor also identified. Fit means your shipping lane is short, your language burden is low, your payment methods work and your category has room. Rank on fit and treat size as a ceiling check rather than a primary criterion.
Assuming a duty rate instead of verifying it
Duty rates are product-specific and origin-specific, and 2025 and 2026 brought substantial changes to US rates in particular. Reading a rate in a blog post, including this one, is not verification. Classification and rate lookups belong on the destination authority’s own system, and for anything ambiguous, a licensed customs broker is worth the consultation fee.
Shipping DAP to protect margin
Billing your customer at the door for duty and tax they did not expect produces refused deliveries, chargebacks and reviews you cannot recover from. DDP costs more per order and converts far better. For consumer sales, treat DDP as the default and DAP as the exception you can justify.
Launching two markets simultaneously
Teams do this to hedge, and it reliably produces two under-resourced launches instead of one good one. Sequence them. Market one produces a playbook, and market two runs on that playbook at a fraction of the cost.
Ignoring the payment layer until launch week
Card penetration varies enormously. In several large European markets, bank transfer and local wallet schemes carry a substantial share of e-commerce volume, and in parts of Latin America and Southeast Asia, cash-adjacent and instant-transfer methods dominate. A checkout that offers only what works at home will convert far below local benchmarks and you will misread that as weak demand.
Treating a marketplace listing as market entry
Listing on a local marketplace is a demand test, not an entry strategy. It tells you whether people buy your product at your price. It does not tell you whether you can operate a returns process, hold local stock, or comply with tax registration when volume crosses a threshold. Use the listing as a probe, and plan the operations separately.
Underestimating the platform work
Multi-currency pricing, tax rules per country, localized checkout, region-specific shipping rules and translated product data all sit on your storefront. Some platforms handle this natively and some fight you. If your current stack cannot express the rules the market requires, that constraint belongs in the selection decision, not in a discovery meeting three weeks before launch. Sellers who hit this wall sometimes conclude the platform is the blocker, which is one of the more common triggers behind a migration from PrestaShop to Shopify or a comparable replatform.
Examples from US retail and e-commerce
Patterns are easier to see at scale, and the large operators have run every version of this experiment publicly.
The cross-border discounters and the limits of a shipping model
The direct-from-origin parcel model built extremely large businesses on a specific set of conditions: low-value parcels, favorable duty treatment and consumers willing to accept long transit times for low prices. When the duty treatment changed, the model’s economics changed with it, and the operators responded by localizing inventory and adjusting which markets they pushed hardest. Their attempted pivot into Latin America, which we examined in the context of Shein and Temu facing entrenched regional incumbents, is a useful reminder that a model tuned for one corridor does not automatically transfer to another.
The lesson for a small seller is not about scale. It is that a market chosen because a specific rule made it cheap is a market you hold only as long as that rule holds. Rules are policy, and policy moves.
US brands entering Europe: the language assumption
American direct-to-consumer brands routinely start European expansion in the UK and Ireland, then treat the Netherlands and the Nordics as a natural second step on the basis of high English proficiency. The demand logic is sound. The operational logic is where it gets interesting: post-Brexit, the UK and the EU are separate customs territories, so a UK-first entry does not produce a playbook that transfers cleanly to the EU. Sellers who understood that upfront chose their sequence deliberately. Sellers who did not discovered it at their second launch.
Canada and Mexico: adjacency is not the same as simplicity
Proximity makes the freight cheap and the transit fast, which is why so many US sellers start there. Neither market is a domestic extension, though. Both have their own tax registration mechanics, French-language requirements apply in Quebec, and duty treatment depends on origin and on whether the goods qualify under the North American trade agreement’s rules of origin. The agreement governing that trade is worth understanding at least at the level of whether your goods qualify, because the answer changes your duty exposure directly.
Reading the aggregate data without overreading it
Public data helps set the frame. The US Census Bureau publishes quarterly e-commerce and trade statistics that give a reliable sense of channel direction, and comparable statistical agencies exist in most destination markets. What none of that data tells you is whether your particular SKU, at your particular price, delivered on your particular timeline, will sell. Use aggregates to eliminate obviously wrong markets and to sanity-check your assumptions, then rely on your own 30 to 50 order test for the decision itself.
Tools, partners and vendors worth knowing
You do not need a large stack for a first market, but you do need four functions covered, and the sequence in which you add them matters.
Duty and landed cost calculation
Several vendors provide HS classification assistance and real-time landed cost quoting that plugs into checkout. For a first market with a narrow catalog, a spreadsheet built from verified rates works fine and forces you to understand the numbers. Automate once your catalog or your market count makes manual maintenance error-prone.
Cross-border shipping and clearance
The major integrators offer DDP services with clearance handled end to end. Specialist cross-border consolidators can be materially cheaper at volume by pooling shipments and clearing them as a single entry. Ask any candidate carrier three specific questions: what the all-in cost per parcel is including disbursement fees, what the returns path looks like and what happens operationally when a shipment is held.
Payments and currency
Your processor needs to support the local methods that matter in the destination and to settle in a currency that does not bleed you on conversion. Look past the headline transaction rate to the cross-border surcharge and the FX spread, which together often exceed the base rate.
Tax registration and compliance
Once you cross a registration threshold, you need a compliance provider or an accountant with genuine experience in that jurisdiction. This is the function most worth paying for rather than improvising, because the failure mode is a retroactive assessment rather than an inconvenience.
Local fulfillment, when the numbers justify it
Third-party logistics providers in-market convert your international parcel problem into a domestic one and make returns workable. The trigger point is usually volume-based: when the per-order saving from domestic shipping and the reduction in returns cost together exceed the fixed cost of holding stock abroad. Model that threshold before you launch, so you recognize it when you reach it rather than debating it for two quarters.
Sequencing the stack
Add landed cost and shipping before launch, payments at launch, tax compliance when you approach a threshold and local fulfillment when volume justifies it. Buying all four upfront is a common way to make a first market look unprofitable when it is merely early. The channel decisions that sit alongside this stack, including which marketplaces to list on and how to structure your own storefront, are covered in our complete guide to selling on global e-commerce marketplaces.
A note on legal, tax and customs advice
Everything above is general information and education about how cross-border selling works. It is not legal, tax or customs advice, and it is not a substitute for professional guidance on your specific situation. Duty rates, tax thresholds, registration triggers and product compliance requirements vary by product, by origin country, by destination and by the exact facts of your business, and they change frequently.
Before you commit inventory or set prices for a new market, the sensible step is to confirm your position with a licensed customs broker, a trade attorney or a tax advisor qualified in the destination jurisdiction. Primary sources are the authorities themselves: US Customs and Border Protection and the Office of the United States Trade Representative for US import matters, the Federal Register for the text of US regulatory changes, the European Commission for EU customs and VAT rules, and HM Revenue and Customs for the UK. Where this article mentions a threshold, a rate or a rule, treat it as a pointer to check rather than a figure to rely on, because current numbers must be verified at the official source.
FAQ
How many markets should I evaluate before choosing one?
Start with a long list of eight to twelve, cut to three to five using hard elimination filters, then score only the survivors in detail. Scoring more than five properly takes longer than the decision is worth, and scoring fewer than three usually means you skipped the elimination step and are just rationalizing a preference you already had.
Is it better to start with a marketplace or my own store abroad?
A marketplace listing is the cheaper way to test demand because it borrows the local audience, trust and payment methods. Your own store gives you margin, customer data and brand control but requires you to solve traffic, localization and payments yourself. Most small sellers test on a marketplace and build their own storefront in that market only once the demand signal is confirmed.
What is a realistic budget for a first international market?
It varies widely by category, but the components are predictable: a test shipment of 30 to 50 orders, localization and translation work, any required registrations, and a working capital buffer for returns and refunds during the learning period. The largest hidden cost is management time, which is rarely budgeted and usually the binding constraint for a small team.
Do I need a local company to sell into another country?
Often no for straightforward direct-to-consumer sales, but it depends entirely on the destination and the product category. Some jurisdictions require a local entity, a fiscal representative or a responsible person for certain regulated products. This is a question for a qualified advisor in that jurisdiction rather than a general rule, because getting it wrong creates liability rather than inconvenience.
How do I find the right HS code for my product?
Destination customs authorities publish searchable tariff schedules, and many offer binding ruling procedures that give you a definitive classification in writing. For products that sit near a classification boundary, and many do, a licensed customs broker is worth the fee, because misclassification can surface as back-duty and penalties long after the shipments cleared.
Should I ship DDP or DAP to consumers?
DDP for consumer sales in almost every case. When the customer is billed for duty and tax at the door under DAP, refused deliveries and chargebacks rise sharply and the reviews are difficult to recover from. Build the duty into your displayed price instead, and price the market on that basis from the start.
How long should I wait before judging whether a market is working?
Give it at least two full inventory cycles or six months, whichever is longer. The first weeks are dominated by launch noise, and the return rate, which is one of the most important signals, only becomes readable after enough orders have passed through the full return window.
What single factor most often makes a first market fail?
An incomplete landed cost model. Sellers price against product cost plus freight, miss the clearance fees, disbursement charges, FX spread and returns cost, and then find that a market with genuine demand generates no margin. The demand was never the problem, and that is why the failure tends to surprise people.
Can I use the same product catalog in every market?
You can, but you usually should not. Heavy items, low-margin items and high-return items behave much worse across a border than they do domestically. Trimming the international assortment to the SKUs with the best weight-to-value ratio and the lowest return rates frequently converts a marginal market into a profitable one.