Cross-border commerce is the fastest way for a US retailer to add a second growth curve once the domestic funnel matures. It is also the fastest way to incinerate working capital if the first international market is picked on instinct. The goal of this guide is simple: give a small or mid-sized e-commerce team a repeatable scoring method to choose first cross border market without guessing, without consultant theatrics, and without copying what a louder brand on LinkedIn happens to be doing.
Most teams fail this decision in one of three predictable ways. They pick the market that feels familiar (Canada or the UK because the language matches), they pick the market with the highest GMV (Germany or Japan because the spreadsheets look impressive), or they pick a market because a friendly 3PL has capacity there. None of these are wrong on their own. None of them are a decision either. They are convenience dressed as strategy.
In short
- Score, do not guess. Use a 6-factor model (demand, payments, logistics, tax, returns, language) with weights tied to your category, then rank candidates.
- Test before you commit. A 90-day controlled pilot with ad spend, a localized PDP, and a real returns path beats six months of desk research.
- Margin lives in payouts, not in revenue. FX spreads, IOSS or sales-tax remittance, and chargeback exposure decide whether the market is profitable.
- Pick a wedge category, not a catalog. The first market should validate one SKU family with strong unit economics, not your full assortment.
- The right answer is rarely the obvious one. Canada and the UK are crowded; Mexico, Australia, the Netherlands and Ireland often look better on a real scorecard.
This article sits inside the wider Understanding global trade for retail and cross-border commerce pillar, which covers the policy, payments and logistics context behind every individual market choice. If you have not read the pillar yet, skim it first; the framework below assumes you understand the basics of Incoterms, landed cost and the OECD-aligned VAT regimes.
Why the first market decision is different from every market after it
The first market sets the operating template. It is where you build the playbook, the spreadsheets, the carrier contracts, the customer-service macros, the returns reverse-logistics flow, and the merchant-of-record relationship you will later copy into market two, market three and market five. A bad first pick does not just lose money. It teaches your team the wrong instincts.
Consider the difference between a US apparel brand expanding to Canada versus the same brand expanding to Mexico. Both look like neighbors on a map. The Canadian pilot teaches the team that cross-border is mostly a duty and a slower carrier transit. The Mexican pilot teaches the team about FX, COD demand, local payment methods (OXXO, SPEI), Spanish-language customer service, and an entirely different returns culture. The Canadian playbook does not generalize. The Mexican playbook does.
That asymmetry is why we recommend teams resist the “let us start somewhere easy” framing. Easy markets validate that the product can be shipped abroad. They rarely validate that you can run a multi-market operation. If your three-year goal is one extra country, pick the easy one. If your three-year goal is six markets, pick the one that forces you to build the muscles you will need later.
The six-factor scorecard for choosing a first market
Below is the model we recommend for retail and direct-to-consumer brands doing under $50M in annual revenue. Each factor is scored 1 to 5; weights are assigned based on your category. Sum the weighted scores and the highest total wins, with one veto rule: any factor scoring a 1 disqualifies the market regardless of total.
| Factor | What it measures | Default weight | When to raise the weight |
|---|---|---|---|
| Demand depth | Search volume, category growth rate, competitor density in the country | 20% | Niche or premium SKUs |
| Payments fit | Local payment methods, card penetration, BNPL, FX spread on payouts | 20% | AOV above $150 |
| Logistics cost | Landed cost as a percentage of selling price, transit time variance | 20% | Bulky or heavy SKUs |
| Tax and compliance | VAT or GST regime, threshold rules, IOSS or merchant-of-record options | 15% | Multi-country expansion plans |
| Returns reality | Return rate norms, reverse logistics cost, consumer protection rules | 15% | Apparel, footwear, electronics |
| Language and CX | Coverage by existing team, translation effort, customer-service tooling | 10% | Complex products or B2B leans |
Weights should add to 100. If you are a beauty brand with an AOV of $42, push demand and payments up and language down. If you are a furniture brand with an AOV of $1,800, logistics dominates everything else. The weighting argument is the most valuable conversation a leadership team can have before any market research begins, because it forces explicit prioritization.
Scoring demand depth without buying a $30,000 report
You do not need a paid research subscription. A workable demand picture can be assembled in a day using free or low-cost tools: Google Trends for relative interest by country, Statista summary pages and the US Census foreign trade data for category-level export volumes, marketplace category browsing on Amazon.de, Amazon.com.au, Mercado Libre and Allegro to count active SKUs, and Ahrefs or Semrush country-level search volume exports for your top 20 commercial keywords.
Score a 5 when you find evidence of an underserved high-intent search cluster (CPC below $1 in a category with growing volume). Score a 1 when the SERP is dominated by 10+ entrenched local competitors and a marketplace giant. Most markets land between 2 and 4; do not over-fit the score, the scorecard is comparative, not absolute.
Scoring payments fit
This is the factor most US teams underestimate. The combination of card mix, BNPL preference, bank-transfer behavior and FX payout cost can swing your margin by 4 to 7 percentage points. We cover this in depth in Cross-border payouts: the hidden friction in your margin, which every founder evaluating a first market should read alongside this scorecard.
A quick proxy: count the local payment methods your current PSP supports in the candidate country. Stripe, Adyen and Checkout.com publish clear coverage matrices. If your PSP supports the top 3 methods, score 4. If you would need to add a regional PSP just to launch, drop the score to 2. If a method like Pix in Brazil or iDEAL in the Netherlands accounts for over 50% of national e-commerce volume and you cannot accept it, that is a 1.
Scoring logistics cost
Build a quick landed-cost spreadsheet for a representative SKU. Include factory or warehouse FOB cost, outbound freight to the destination country (DHL, FedEx and a local consolidator quote), customs duty at the HS code, VAT or GST at the destination rate, last-mile delivery, fuel and remote-area surcharges, and a reverse-logistics reserve at 1.5x the forward cost.
Express your landed cost as a percentage of the local selling price. Anything under 35% is a 5. Anything over 60% is a 1. The boring middle (40 to 55%) is where most first-time cross-border programs live, and where pricing discipline matters more than any other lever.
Scoring tax and compliance
The EU’s IOSS scheme, the UK’s overseas seller VAT regime, Australia’s GST low-value-imported-goods rules, Canada’s GST/HST nonresident registration, and Mexico’s IVA rules all differ in ways that matter to a small team. Two questions decide the score: can a merchant-of-record (Global-e, Reach, ESW) cover this market for you, and what is the long-run cost of that coverage as you scale?
If a merchant-of-record covers it cleanly at 3 to 5% of revenue and you plan to use them for the first 18 months, score 4. If you need to register directly from day one (and absorb the accounting cost), score 2. Our detailed explainer on IOSS, OSS and the EU VAT rules for cross-border sellers goes deeper on the EU-specific mechanics.
Scoring returns reality
Returns are where cross-border margin goes to die quietly. EU consumers expect a 14-day no-questions-asked return window under the Consumer Rights Directive. UK consumers expect free returns on apparel from any serious brand. Mexican consumers return less often but expect longer delivery windows. Japanese consumers return almost never but expect packaging perfection.
Score 5 if you can plug into a regional returns aggregator (ReturnGo, Loop, ZigZag) that consolidates at a local hub. Score 1 if every return has to come back to the United States, because that single fact will sink your unit economics on any apparel SKU.
Scoring language and CX
Translation cost is trivial in 2026. The hidden cost is customer service. A 9-to-5 US team cannot answer European or Australian queries on local time, and machine-translated replies destroy CSAT once the conversation passes turn two. Consider whether you can hire a single Lisbon-based or Mexico-City-based contractor for 20 hours a week, and whether your help-desk software (Gorgias, Zendesk, Front) supports the local-language macros you need.
Working an example: a $12M home-goods brand picks its first market
Take a US home-goods brand selling ceramics and small kitchenware with an AOV of $95 and a 38% gross margin. The team has a long list of candidate markets but, like every team, can only execute one. Below is the scorecard applied to four realistic candidates.
| Factor (weight) | Canada | UK | Australia | Mexico |
|---|---|---|---|---|
| Demand (20%) | 4 | 4 | 3 | 3 |
| Payments (20%) | 5 | 4 | 4 | 2 |
| Logistics (20%) | 4 | 2 | 2 | 4 |
| Tax (15%) | 3 | 3 | 4 | 2 |
| Returns (15%) | 3 | 2 | 3 | 3 |
| Language (10%) | 5 | 5 | 5 | 3 |
| Weighted total | 3.95 | 3.25 | 3.25 | 2.85 |
Canada wins on this scorecard, which is exactly the unsurprising answer. But notice that the UK and Australia scored equally and both have logistics scores of 2, meaning a 60+% landed cost. For this specific brand, that is a veto-grade signal: the UK and Australia are off the list not because they are bad markets, but because the SKU cube is too heavy to make sense from a US warehouse without setting up a 3PL on the destination side.
The non-obvious learning is that the team’s instinct to “do the UK because language is easy” would have been the worst commercial choice on the list. The scorecard caught it. That is the entire point of the exercise.
Common mistakes teams make when picking the first market
Mistake 1: optimizing for “first sale” not “first hundred”
A pilot that produces 10 cross-border orders is not a market entry; it is a curiosity. A real pilot needs to produce enough volume to surface returns behavior, edge-case shipping issues, payment failures, customer-service patterns and chargeback rates. Aim for at least 200 paid orders in the first 90 days before declaring a verdict on a market.
Mistake 2: relying on US ad creative
The same Meta or TikTok creative that wins in Iowa rarely wins in Berlin or Sydney. Voiceover accent, on-screen pricing format, social proof (Trustpilot in the UK, ekomi in Germany, ProductReview in Australia) and even color contrast in mobile-first markets all matter. Budget 15 to 20% of pilot spend for creative localization, not just translation.
Mistake 3: shipping the full catalog
If you sell 400 SKUs, do not launch with 400 SKUs in a new country. Pick a wedge of 15 to 40 hero SKUs that share packaging dimensions, ship density and HS code. Constraining the assortment makes carrier negotiation, customs declarations and returns processing dramatically simpler.
Mistake 4: assuming the marketplace will do the work
Listing on Amazon.de or eBay.co.uk is not market entry; it is a sales channel test. It will tell you whether your product sells abroad. It will not tell you whether your direct brand can build a customer base abroad, which is a different question with different answers. Use marketplaces as a demand probe, not as a strategic conclusion. Our deep dive on Tools and vendors for ebay in 2026 is a good companion read if you are using eBay as part of that probe.
Mistake 5: ignoring the calendar
Black Friday matters in the US, the UK and Germany. It barely matters in Japan. Singles’ Day matters in China and increasingly in Southeast Asia. Boxing Day matters in Canada, the UK and Australia. Aligning your launch calendar with the destination’s commercial peaks (and not just the US Q4) can double the ROI of the first six months of ad spend.
Mistake 6: treating the merchant-of-record as a black box
If you outsource compliance to a merchant-of-record, you still need to understand what they are doing. When VAT rates change, when an OSS threshold shifts, or when a chargeback dispute hits, you need to know which contract terms protect you and which expose you. Sign the MoR contract with your finance lead in the room, not just your ops lead.
Five real US retail and e-commerce examples
- Allbirds in the UK (2018): A clean first-market choice driven by AOV fit (the UK supports premium pricing), a brand-led launch with a flagship Covent Garden store, and language alignment. Worked because the product travels well; the limitation was that the UK pilot did not generalize easily into mainland Europe.
- Glossier in Canada (2017) and the UK (2017): Two simultaneous first markets, which is rare and only works when the brand is already global by reputation. Mostly successful, but the team has since reported that running both markets in parallel doubled the operational complexity in year one.
- Warby Parker in Canada (2014): A textbook conservative pick. The Canadian launch was a logistics extension more than a market entry, and the lessons translated less than the team initially expected when later markets were considered.
- Dollar Shave Club in Australia (2015): A counterintuitive pick driven by subscription density and high credit-card penetration. Worked because the Australian market is small enough to test a model and large enough to be material.
- Casper in Germany (2017): Driven by a clear bedding category and a willingness to invest in localized creative. Eventually wound down, but the lesson is instructive: high logistics costs on bulky goods can defeat strong demand signals.
The pattern across these examples is simple. The brands that picked first markets where the playbook generalized (Glossier, Allbirds in some respects) built faster international engines than the brands that picked easy markets for convenience. Your goal is to be in the first group.
The 90-day pilot template
Once the scorecard picks the market, run a structured pilot. Below is a template we recommend for teams with under $5M in cross-border-pilot budget. The goal of the pilot is not revenue; it is data quality.
| Week | Workstream | Deliverable |
|---|---|---|
| 1-2 | Setup | Localized PDP, currency display, payment methods live, MoR contract signed |
| 3-4 | Soft launch | Paid ads at 25% of planned spend, 50 orders target, returns flow tested |
| 5-8 | Scale | Full ad spend, creative iteration, customer-service SLA measured |
| 9-10 | Stress test | Peak-traffic simulation, returns surge test, chargeback drill |
| 11-12 | Decide | Scorecard review, GMV vs. plan, contribution margin assessment |
At week 13, the team meets and answers three questions: did the unit economics work, did the operations scale without burning out the US team, and is there a credible path to 5x revenue in the next 12 months. If two of three are “yes,” scale. If one of three is “yes,” redesign and re-pilot. If none, stop and pick a different market.
Tools, partners and vendors worth knowing
The cross-border ecosystem is mature in 2026, and there is no shortage of capable vendors. The shortlist below covers the seven categories every first-market team needs to staff before launch.
- Merchant-of-record: Global-e, Reach, ESW, Digital River. Choose based on the markets they cover natively and the contract terms around chargebacks.
- Payments: Stripe, Adyen, Checkout.com. All three cover most useful first markets. Adyen tends to be stronger in Europe and Australia; Stripe is the easiest to bolt on alongside a US setup.
- Logistics: DHL Express (premium), FedEx International Economy (value), Asendia (consolidation), local last-mile partners like Royal Mail, Australia Post, FedEx, Sendle, Estafeta, DPD.
- Translation and localization: Smartling, Phrase, Crowdin, and for the leanest setup a curated network of native-speaking freelancers via the localization marketplace of your choice.
- Returns: ReturnGo, Loop Returns, ZigZag Global, narvar Concierge. The European players (ZigZag, ReBound) tend to have the deepest local consolidation networks.
- Tax compliance: Avalara, TaxJar, Vertex, Sovos. Most US teams already use Avalara for state sales tax; extending to VAT and GST is usually easier than starting from scratch.
- Analytics and FX visibility: Mode, Looker Studio dashboards stitching ad spend, GMV, payouts and FX. The FX visibility is the most commonly missed piece; budget a half-day to build it before launch.
How to sequence markets two through five once the first one is working
The first market is a single decision; the path beyond it is a portfolio. Once the pilot is producing positive contribution margin at week 13, the natural impulse is to “do the same thing again in a different country.” That impulse is half right. The second market should reuse the operating template (carriers, MoR contracts, returns aggregator, payment stack) but should consciously stretch one new muscle: a different language, a different payment culture, a different consumer-protection regime, or a different demand season.
A useful rule of thumb is the “one new variable” principle. If your first market was Canada, your second should add language complexity (Mexico) or payment complexity (the Netherlands), but not both at once. If your first market was the UK, your second should add a non-Latin-character market (Japan) or a high-FX-volatility market (Brazil) only if you have the appetite to invest in a second operating playbook. Stacking two new variables on the second market roughly triples the operational risk; teams that try it often end up doing neither market well.
By market five, you should have a clear internal vocabulary for what your operating template can and cannot absorb. Most teams discover that the template handles 3 to 4 markets gracefully, then breaks somewhere on the fifth as the customer-service organization can no longer be staffed from a single time zone. That is the natural inflection point at which a regional ops lead in either Europe or APAC starts to make economic sense.
How this connects back to the wider trade story
Picking the first market is one chapter inside a longer story about how US retail brands build a global footprint. The decisions you make on market one constrain the carriers, payment processors, MoR partners and operational rhythms you can use in market two. Returning to the global trade pillar after running a pilot is one of the highest-leverage things a leadership team can do, because the policy and macro context that felt abstract before the pilot will suddenly feel actionable.
One closing point. The single biggest predictor of success in cross-border programs we have observed is not which market a brand picked first. It is whether the team treated the first market as a learning system or as a revenue project. The teams that treated it as learning built durable, profitable global businesses. The teams that treated it as a quarterly revenue goal built tactical, fragile ones that collapsed when ad costs rose or freight rates moved.
FAQ
Should I always start with Canada or the UK?
No. Canada is often a strong default because of language, currency familiarity and logistics, but the UK can be a poor first choice for bulky-goods brands due to landed cost. Run the scorecard against your category before defaulting to either.
How much should I budget for a 90-day pilot?
For a small or mid-sized brand, plan on $80,000 to $250,000 covering setup, ad spend, inventory pre-positioning, MoR fees and a returns reserve. Lower budgets are possible if you start on a marketplace, but a direct-brand pilot below $80,000 rarely produces enough data to make a confident scale decision.
Can I use a merchant-of-record forever or do I need to switch eventually?
You can use an MoR indefinitely, and many brands above $50M in cross-border revenue still do. The economics flip somewhere between 5 and 12% of GMV depending on category margin; below the flip point the MoR is a clear net positive, above it you may save money by going direct.
What KPI should I watch most closely in the pilot?
Contribution margin per order after FX, returns, refunds, taxes and chargebacks. Revenue and conversion rate are vanity metrics in a cross-border pilot; contribution margin is the only number that tells you whether the market is real.
Do I need a local entity in the first market?
Usually not for the first 18 months. A merchant-of-record removes the need for an entity in most jurisdictions. Set up a local entity only when the MoR fee starts to materially exceed what a local accountant and VAT registration would cost.
How do I handle customer service across time zones?
Hire one part-time contractor in the destination time zone before paid ads turn on. Two hours of live human response in local time outperforms 24-hour AI coverage in customer satisfaction scores, and the cost difference is small at pilot scale.
What is the most common single-point-of-failure in a first market?
Returns. Specifically, the absence of a local returns aggregator. Brands that try to route returns back to the US for cost reasons see customer-service tickets, refund delays and bad reviews compound until growth stops. Solve returns first, before scaling spend.
How do I know when to stop and pick a different market?
If after 90 days your contribution margin per order is negative, your CSAT in the destination is below your US baseline by more than 10 points, or your team is spending more than 25% of its time on cross-border firefighting, stop. Re-score the candidates and pick again. There is no shame in killing a pilot; there is shame in scaling a broken one.