International ecommerce expansion: the complete guide

Pick a single country first, and pick it from your own data rather than from market size. You almost certainly already have signals about where demand exists, and following those is far cheaper than guessing.

Start with the traffic you already get. Look at which countries send you visitors, which ones already produce orders despite you doing nothing for them, and which ones abandon at the shipping step. A country that buys from you already, with an unhelpful delivery promise and prices in the wrong currency, is telling you something worth listening to.

Then look at the practical filters. Shared or widely spoken language reduces the localisation cost sharply. Geographic proximity or an existing trade arrangement usually means cheaper, faster shipping and simpler customs. Similar consumer protection expectations mean your returns policy needs less rework. A market that scores well on all three is a much easier first step than a larger one that scores badly.

Check the category reality. Is your product type already sold widely there. Are there local suppliers with a delivery advantage you cannot match. Are there rules about your product category, such as labelling, ingredients, electrical standards or age restrictions, that make it unsuitable. Some categories cannot be shipped across certain borders at all, and finding that out before you build the site is preferable.

Finally, size the opportunity honestly at your own scale. The question is not how big the market is, it is how many orders a month you could realistically win in the first year and whether that number justifies the setup work and the ongoing operational load. Expanding into a market you cannot service properly costs more than staying home.

Write the choice down with the reasons. When the first few months are slow, that written reasoning is what stops you from abandoning a good decision or persisting with a bad one.

Calculate what it actually costs to get one order into a customer's hands in that country, then set the price. Doing this the other way round is how stores end up shipping at a loss and discovering it a quarter later.

Landed cost is the product cost plus international shipping, plus any duty and import taxes, plus the carrier's handling or clearance fees, plus payment processing which is often higher on cross border transactions, plus the currency conversion cost, plus the cost of a return from that country. Add them all before you compare to the price you plan to charge.

The biggest decision is who pays the duty and import tax. Delivering with those charges paid means you collect them at checkout and the buyer receives the parcel with nothing more to pay. Delivering without means the carrier collects from the customer on arrival, often with a handling fee on top. The second option looks cheaper and it produces refused parcels, angry messages and poor reviews, because the buyer did not expect a bill. For most consumer products, charging the full amount at checkout and delivering with nothing to pay on arrival is the better commercial choice, even though the displayed price is higher.

Work out the threshold effects. Many countries apply different rules below and above a value threshold, so the same basket can be treated differently depending on its size. If a threshold sits near your typical order value, it is worth knowing where it is before you set a free shipping line.

Then decide the price. Converting your home price at the day's exchange rate produces odd looking numbers and does not account for the extra costs. Set a local price that looks like a normal price in that country and that covers the landed cost with the margin you need. Review it when exchange rates or carrier rates move rather than leaving it fixed for years.

Say the total clearly. A price that turns into a bigger number at checkout is the fastest way to lose an international shopper.

Choose how the international version of your store will be addressed before you build anything, because the URL structure is difficult and disruptive to change once search engines and customers have found it.

There are three common approaches. A country specific domain signals the market clearly and is unambiguous, but it means running and building authority for a separate site. A subfolder on your existing domain keeps everything under one site and benefits from the authority you already have, and it is usually the simplest option for a first expansion. A subdomain sits between the two. For most stores opening one additional market, a subfolder is the practical choice.

Whatever you choose, the different versions have to tell search engines about each other. Language and region annotations on each page, pointing at every other version including itself, are what stop the wrong version ranking in the wrong place and reduce the risk of your versions being treated as duplicates. Get these right at launch, because errors here are common and quietly costly.

Avoid the automatic redirect trap. Sending every visitor to a version based on where they appear to be, with no way out, frustrates people who want a different version and can prevent search engines from seeing your other pages. Offer a clear, dismissible suggestion and let people choose.

Decide what actually differs between versions. If only the currency, the shipping information and a few phrases change, much of the content is shared, and you need to be careful that near identical pages are handled properly. If prices, assortment, sizing conventions and legal information all differ, the versions are genuinely different pages and should be treated as such.

Finally, plan the technical basics: local currency display, a way to switch versions in the header or footer, correct sitemap entries per version, and search console properties set up for each so you can see how each market is performing separately.

Display prices in the local currency and support the ways people in that country normally pay. These two things move international conversion more than almost anything else on the page, and both are usually settings rather than development work.

Currency first. A shopper who sees a foreign currency has to do mental arithmetic and wonder what their bank will charge. Showing the local currency removes that friction. Make sure the currency shown on the product page is the currency charged at checkout, because a price that changes at the payment step causes abandonment. It also matters for shopping listings, where the currency in your product data has to match the currency on the landing page.

Then payment methods. Card penetration and preferred methods vary a lot by country. Some markets are dominated by bank transfer style methods, some by local wallets, some by buy now pay later, and some by cash on delivery. Offering only the methods that work at home leaves orders on the table. Check what the payment providers you already use support in that market before assuming you need a new one.

Check the checkout details that break across borders. Address formats differ, postal code rules differ, and phone number fields that expect a home country format will reject valid numbers. Test the whole checkout with a real local address, not a made up one.

Watch the fees. Cross border card transactions and currency conversion both cost more than domestic ones, and those costs belong in the landed cost calculation from the previous chapter. If you sell at volume in a market, a local payment arrangement can reduce them, but that is an optimisation for later rather than a launch requirement.

Finally, make the tax presentation match local expectation. In some markets shoppers expect to see prices with tax included, and showing prices without it makes you look more expensive at the payment step than the competition looked on the shelf.

Publish a delivery time and a returns route you can deliver on every time, then build the operation to match. International buyers are more nervous than domestic ones, and the promise is the main thing that reassures them.

Get real transit times from the carriers you will actually use, including customs clearance, and publish a range rather than an optimistic single figure. Then look at what that range does to your conversion rate. If your delivery estimate is far longer than what local sellers offer, that gap is your main commercial problem in the market, and no amount of advertising will fix it.

Decide the fulfilment model. Shipping every order individually from home is the simplest way to start and the slowest. Bulk shipping stock to a fulfilment partner in or near the market costs more to set up and transforms the delivery promise, the shipping cost per order and often the returns experience. Start with the simple option, and treat the switch as a decision to make once the market proves itself.

Returns need a plan before launch, not after the first request. Asking a customer to ship an item back across a border at their own cost is a policy that suppresses sales and generates complaints. A local return address, a consolidated return service, or a returns partner is what makes the policy credible. In some markets, consumer law sets minimum rights around cancellation and returns, so check what applies rather than assuming your home policy transfers.

Decide what happens to returned stock. Bringing it home may cost more than the item is worth, so a local resale, refurbishment or donation route is worth arranging in advance.

Finally, publish the details in plain language on the product page and the checkout: the delivery range, the cost, whether duties are included, and how to return something. Uncertainty is what stops international orders, and every uncertainty you remove is worth more than a discount.

Deal with the legal and tax side before you spend money bringing traffic, because the cost of getting it wrong is disruption to the whole market rather than a marketing setback. This is the part to get professional advice on, and this chapter is a checklist of what to ask about rather than advice itself.

Ask about registration thresholds. Many countries require a seller to register for local sales tax once they pass a level of sales, and some require registration from the first sale for distance selling. The answer determines whether you need to collect tax at checkout and file returns there, and it changes your pricing.

Ask about import arrangements. There are usually schemes designed to simplify low value imports, and using one changes how tax is collected and how parcels clear customs. Your carrier or a customs broker will know what applies to your product and your volumes.

Get your product data right for customs. Correct commodity codes, accurate country of origin, and accurate declared values on every parcel. Wrong codes cause delays, unexpected charges and returns, and they are entirely avoidable.

Check product level rules. Labelling requirements, language requirements on packaging, safety marks, ingredient restrictions, electrical plug standards, age restrictions and category specific advertising rules all vary. Some of these affect whether you can sell the product at all, and some affect whether you can advertise it on a given platform in that country.

Check the consumer side too: required information on the website, cancellation rights, warranty periods and how complaints must be handled. Some of this has to appear on your site, in the local language, before you take the first order.

Do this work first, and keep a record of what you established and when. Rules change, and a documented position is much easier to update than one held in someone's memory.

Translate everything a buyer reads, and go further than translation on the parts that decide the sale. Machine translated product copy with home country conventions left in place reads as careless, and carelessness is expensive when nobody has heard of you.

Start with the details that are wrong rather than merely foreign. Sizing conventions, measurement units, date formats, address formats, clothing and shoe size systems, electrical specifications, and any reference to a home country service or holiday. These are the errors that make a shopper close the tab, because they suggest the product may not suit them at all.

Then the trust content. Delivery information, returns policy, contact details, tax treatment and any legally required notices should be written in the local language and reflect local expectations. If you can offer a local contact route, say so.

Then the persuasion. Translated marketing copy tends to be flat, because the phrasing that works in one language rarely maps directly. Where a market matters, have a native speaker write rather than translate the important pages: the homepage, the main category pages and your best selling products. Elsewhere, careful translation with a human review is usually enough.

Search terms need the same treatment. People search using their own words, and the direct translation of your product category may not be the term used locally. Redo your keyword research in the target language rather than translating your existing list, or you will optimise pages for phrases nobody types.

Images and proof matter too. Photography featuring recognisable local settings and reviews from customers in that market do more for credibility than a translated testimonial from home. Early on you will not have local reviews, which is a good reason to gather them deliberately from the first orders.

Finally, check the small things: currency symbols in the right position, correct decimal separators, and no leftover home country phone numbers in the footer.

Give each market its own product data, targeting and reporting, because a single blended setup hides which country is working and often breaks the rules the platforms apply to shopping listings.

For shopping listings, the product data has to match the landing page for that country. That means the currency, the price including any tax presentation the market expects, availability, and the shipping cost and delivery time configured for that country. A mismatch between the data and the page is one of the most common reasons items get disapproved, and it becomes far more likely when you add a market and forget to update the data.

Decide whether you run one product data source with country specific overrides or separate sources per country. Either can work. What matters is that the price, currency and link for each country are correct, and that a change to your pricing in one market cannot silently break another.

Configure shipping and tax settings per country in the merchant account so the total shown to shoppers is right. If you deliver with duties paid, that has to be reflected, because a shopper who sees one figure in a listing and a larger one at checkout will leave.

On the advertising side, keep campaigns separate by country. Budgets, bids, seasonality and competition all differ, and combined campaigns will quietly spend most of the money in whichever market is cheapest rather than the one you are trying to build. Separate campaigns also make it possible to pause a market cleanly if it is not working.

Set up analytics so you can read the market on its own: traffic, conversion rate, average order, return rate and contribution, all filtered to that country. If you cannot see those five numbers for the new market separately, you cannot make a fair decision about it later.

Launch with a small, focused amount of spend on the products most likely to work, and make sure someone can answer a question the same day. A new market has no reviews, no brand recognition and no word of mouth, so it needs more support per order than your home market does.

Start with intent. Search and shopping placements reach people who are already looking for what you sell, which is a fairer test of whether the product and the price work in that market than social advertising to people who have never heard of you. Once you know the offer converts, broadening into discovery channels is a reasonable next step.

Pick a small set of products for the launch rather than the full catalogue. Your best sellers, with the strongest images and the most reviews, priced properly for the market. A narrow launch is easier to service, easier to read, and cheaper to abandon if the answer is no.

Expect the numbers to look worse than at home at first. There is no returning customer base, no email list and no brand searches, so the blended figures are unflattering. Judge the launch on the same measures you would use for cold traffic at home, not on your overall store average.

Staff the questions. A shopper in a new market asks about delivery time, customs charges, returns and sizing, and the answer they get decides whether they order. Publish those answers prominently, and make sure someone responds within a working day in that market's time zone. Where you can, offer a contact route that does not require an international call.

Finally, collect the early feedback deliberately. Ask the first customers how the delivery went, whether anything was unclear and whether they were charged anything on arrival. Those answers will find the operational problems faster than any dashboard.

Give the market a defined trial with numbers agreed in advance, then make a decision rather than letting it drift. International expansion consumes attention across the whole business, so an undecided market is expensive even when it is small.

Measure four things. Contribution per order after landed cost, duties, payment fees and returns, which tells you whether the unit economics work. Conversion rate compared with your home market on similar traffic, which tells you whether the page, price and delivery promise are credible there. Return rate, which tells you whether sizing, expectations or shipping damage are a problem. And repeat purchase rate over time, which tells you whether you are building a business in that market or just buying single orders.

Set the trial length to fit your purchase cycle. A market selling a consumable needs long enough for a second order to be possible. A market selling a considered purchase needs long enough for people to decide. A quarter is often too short to judge repeat behaviour, and a year is usually longer than you need to judge the basics.

When you review, separate the problems. A market with good conversion and poor contribution is a pricing or shipping problem. A market with poor conversion and good traffic is a page, price or delivery promise problem. A market with high returns is usually a product information problem. Each has a different fix, and only one of them is a reason to leave.

If it works, deepen before you widen. Local fulfilment, a local return route, local language content written rather than translated, and reviews from local customers all improve a market you already have, and they usually pay back faster than opening a second country.

If it does not, close it properly. Stop advertising, keep the site version live long enough to serve existing customers, honour the returns window, and write down what you learned. The next market will go faster for it.