A convenient myth surrounds expansion: that foreign demand is “already visible” in your own analytics if you just look for it. A shop without foreign-language versions has almost no such signals and cannot have them. That is no reason to give up on expansion; it is a reason to approach it in the right order: your own data selects the products, external tools check demand and prices, and data about the market itself is only created by a cheap, properly measured test.
No illusions
The signals myth: what a shop won’t see in its data
Three reasons why you won’t find foreign demand in your own reports:
Google matches language, not intentions
A Czech user searches in Czech and gets Czech results. A shop without a Czech language version is not even in the running, so it collects no impressions or rankings there; the lack of a signal reflects the missing language version, not a lack of demand.
The checkout stops the rest
Even a customer who somehow lands on the domestic site sees prices in a foreign currency, domestic-only delivery and terms in a foreign language. The conversion of such traffic reflects the barriers, not the market.
Occasional foreign orders are noise
They do happen: expats living abroad, purchases by model name, customers near the border. A handful of such orders is an anecdote, and a market ranking must not be built on it.
A conclusion that changes the plan
Before the test, no data about your foreign market exists; the test is your first purchase of it. The test budget is the price of the first real figures: conversion, basket value and margin after delivery in that market. Everything before it is preparation for the purchase, so that you don’t overpay.
Your own data
What your data really shows: products that can bear shipping
Your own data won’t choose the country, but it will reliably choose the ammunition for the test. A product is suitable for a foreign debut when its margin can absorb cross-border delivery costs, shipping is not a gamble (weight, size, fragility) and the offer gives someone a reason to choose it over a local one. An example from a garden shop (illustrative data):
| Category | Margin | Shipping as % of basket | Customer’s reason to choose | Verdict |
|---|---|---|---|---|
| Hand tools (own brand) | 52% | 9% | unique: no local equivalent | test it |
| Irrigation systems | 38% | 7% | comparable goods: price decides | conditional: check local prices first |
| Ceramic planters | 44% | 28% | heavy and fragile: costly shipping, risk of damage | not for the start |
This calculation rests entirely on data the shop already holds: margin per product (where to get it: the guide to margin in e-commerce), sales history and carrier price lists. Rule of thumb: the test should include categories where the margin after foreign delivery costs still looks like a business, not the ones that sell best at home.
External research
Demand and prices are checked outside your own analytics
If your own data says nothing about foreign demand, ask the sources that can see it. All three are available without spending a cent on media:
- Google Keyword Planner set to the target country and language: search volumes for your category phrases (translated properly, not word for word with a machine translator) show whether demand exists at all and how expensive clicks are.
- Google Trends for comparing countries and seasonality: the same category can follow a different calendar in two countries.
- Local price comparison sites and marketplaces: half an hour of browsing competitor prices answers a question more important than market size: whether your price plus delivery fits on the local shelf at all.
These three sources, together with logistics price lists, produce a shortlist of countries: the winner is usually the market that is logistically close, with prices that leave room for margin, not the largest one. Market size does not pay for shipping.
Beyond the data
What sales data won’t tell you: the cost of entry
Demand research is half of the preparation; the other half is the cost of entry, which has to be calculated before the test starts:
Cross-border logistics
Delivery rates to the test country and the real cost of handling returns from abroad. This cost comes out of the margin on every order and can decide whether low-value baskets are profitable.
Local payment methods
Every market has its own payment habits; missing them lowers conversion regardless of demand. Cards are enough for a test, but not for scaling.
Language: feed, shop, customer service
The order matters: for a test, a translated feed and the test category pages are enough. Full shop translation and customer service in the customer’s language are expenses for after a positive test.
Prices, currency and VAT
Prices in the local currency, VAT of the consumer’s country and price positioning against local competitors. A simple exchange-rate conversion is not enough to make a price look credible.
Protocol
Testing a market on a small budget: step by step
Choose test categories, not the whole catalogue
One or two categories from the ammunition table: a margin that absorbs the delivery cost plus a reason for the choice. A properly translated product feed (titles and attributes following the rules in the guide to Google feed fields), because the feed is what will do the selling.
Launch product campaigns for a single country
Product campaigns (Shopping / Performance Max) for the test country, with a separate budget and separate results accounting. No mixing with domestic campaigns.
Measure the result on margin after delivery
Paid orders, product margin minus the cost of cross-border delivery. Revenue without this deduction inflates the test result exactly where it hurts most.
Compare with the home market
Conversion, basket value and acquisition cost against the home market. The test runs for a full purchase cycle of the category; a week is too short for anything but noise.
Decide using thresholds set before the test
For example: conversion at least half of the domestic rate and a positive margin after delivery = scale up (translations, local payments); below the thresholds = close the test and return to the ranking.
Order of investment
Demand first, infrastructure second. Full shop translation, local payments and customer service in the customer’s language are expenses for after a positive test. The reverse order freezes cash in a market whose demand nobody has confirmed.
DataOrganizer
How DataOrganizer evaluates a market test
DataOrganizer comes into play at the two moments when your own data matters. Before the test: margin per product and sales history build the ammunition table. During the test: country campaigns sit in the model like any other campaigns, and orders from the shop system carry the delivery country, so evaluating the test (paid orders, margin after delivery cost, comparison with the home market) fits into a single conversation:
Six weeks of testing, two categories:
| Test category | Spend | CZ orders | Margin after delivery | After ads |
|---|---|---|---|---|
| Hand tools (own brand) | €1,200 | 86 | €1,950 | +€750 |
| Irrigation systems | €900 | 31 | €422.50 | −€477.50 |
Own-brand tools passed the threshold: conversion at 82% of the domestic rate and a positive result already during the test, despite an acquisition cost 12% higher. Irrigation did not: conversion at 41% of the domestic rate and a loss after advertising, which matches the price research (the local shelf is cheaper). Recommendation: scale the tools (full category translation, local payments) and either close the irrigation test or come back with a different price.
Methodological note: the figures in the examples are illustrative. The test is evaluated on the delivery country from shop system orders; demand and price research before the test is done in external tools (Keyword Planner, Trends, local price comparison sites), because no shop’s analytics contains this data. The test runs for at least one full purchase cycle of the category.
Key takeaway
You choose products from your own data, check demand externally and get to know the market through a test. This order protects against two mistakes at once: looking for signals that are not there, and translating the shop for a market nobody has confirmed.
FAQ
Frequently asked questions about international expansion
Can foreign potential be seen in a domestic shop’s data?
Not directly: without a foreign-language version, the shop does not appear for local searches, and occasional foreign orders (expatriates, searches by model name) are noise. Your own data does something else: it identifies products that can carry the shipping cost, and it evaluates the test once it starts.
Which country should you choose first?
Based on the numbers: cost and time of delivery and returns handling, demand and price levels from external sources (Keyword Planner for the language and country, Trends, local price comparison sites), and the language barrier in customer service. The winner is usually the market that is logistically close, with prices that leave room for margin; market size does not pay for shipping.
How can you test a market on a small budget?
A feed for one or two categories in the country’s language, product campaigns for a single market, results measured on paid orders and margin after delivery cost, conversion and basket value compared with Poland, and a decision based on thresholds set before the test.
When should the whole shop be translated?
After a positive test: the test campaigns deliver orders, the margin after delivery cost is positive and the acquisition cost is not wildly out of line with the domestic one. The order: full translation of the test categories, local payment methods and a delivery price list, then the rest of the catalogue and customer service in the customer’s language.
How does DataOrganizer help?
Before the test: margin per product and sales history build a table of products that can carry the shipping cost. During the test: the test campaigns sit in the model like any others and orders carry the delivery country, so settling the test on margin after delivery cost takes just a few questions in the chat. Demand research before the test is done in Google’s external tools.
Cheat sheet for the whole guide
Expansion without illusions in three moves
1. Choose your ammunition from your own data. A margin that absorbs the cost of foreign delivery, shipping without surprises and a reason for the customer to choose you over a local shop: these are the criteria for selecting products for the test.
2. Check demand and prices externally. Keyword Planner for the language and country, Trends, local price comparison sites: the shortlist of countries comes from logistics and prices, not from signals a domestic shop does not have.
3. Test narrowly and count honestly. Categories instead of the whole catalogue, one country, results measured on paid orders with the delivery country and margin after shipping cost, thresholds set before launch. DataOrganizer settles such a test like any other campaign, alongside domestic results, so the decision to scale is based on numbers.
DataOrganizer · market test settled on margin
Connect your shop and ad platforms: test campaign spend is set against orders by delivery country and margin after shipping cost, alongside domestic results. Demo account available instantly; you pay only for connected sources.