One campaign structure for every market

Expanding into another country rarely looks like a project. It creeps up instead: the first orders arrive from a neighbouring country, so an account is opened for it. Then another, because a supplier called. Then a third, because it is "basically the same market". Each one is built by whoever had time, using whatever they had just learned.

By the fifth market it starts to grind. Nobody can answer which market earns the most, because every account has a different shape and different naming. A change that worked at home has to be repeated by hand five times. And the sixth market is built from scratch again.

This article is about the way out: what every country should have in common, what has to stay different, and how markets can then be compared side by side.

How it shows up

Before building anything, it is worth naming what actually hurts about scattered accounts. Most of it comes down to four things:

• Markets cannot be compared. In one country brand sits in its own campaign, in another it is mixed in with everything else. The average cost of sale of the two accounts therefore means different things, and putting them side by side is misleading.
• Every change is made as many times as you have markets. A new negative keyword, a new label, a budget change before the season. At five markets it is doable. At thirteen it simply stops happening.
• A new market is built from scratch. And because someone else builds it, it ends up with yet another shape. The problem grows on its own.
• Budget cannot be moved. When you don't know which market still has headroom and which is saturated, you can't move money to where it would do more.

Notice that none of these is about campaign performance. They are about clarity. And clarity is the first thing expansion costs you.

The market template: what should be the same everywhere

What should stay the same is what gets named and measured. Four things, specifically.

1. The same set of campaigns

Every country gets the same skeleton. For an online shop that usually comes out as three:

• Brand — searches for the brand name, always separate. Not for performance, but so the report can be read at all; we cover why in brand vs non-brand.
• Search — product searches in the market's own language. Not a translated home-market keyword, but what people in that country actually search for.
• Shopping from the product feed — for most shops this is what carries the volume.

Three or four campaigns per market sounds thin, and that is deliberate. The same rule applies as inside a single account: fine-grained splitting starves the automated strategies of the conversion data they learn from. We go through it in Google Ads account structure, and on a smaller market it matters twice as much — there simply aren't enough conversions to split into ten piles.

2. The same naming convention

Something along the lines of [channel] | [brand/nonbrand] | [category] | [market]. Which one you pick barely matters. What matters is that it is consistent, because every report and filter rests on it. When a campaign is called Search_UK_brand in one country and Brand – AT (new) in another, there is no single filter that covers both.

3. The same labels

Campaign labels by margin or business priority, named identically in every country. That is what lets you filter a report by the business rather than by campaign names.

4. The same metric definitions

This one is the quietest and the most treacherous. If one country's cost of sale is calculated from ad-platform conversions and another's from real orders in the shop, the numbers look comparable side by side but are not. Pick one definition for all markets and hold to it; what each metric means is covered in ROAS, CPA, CAC and LTV.

What has to stay different

A shared template does not mean identical settings. These are steered and judged per market, because the market dictates them, not your methodology:

It pays to be careful with returns and shipping in particular. A market where people return a third of the goods and shipping costs twice as much needs a stricter target, even though it looks just as good as the home market inside the ad platform. How to work out margin by category is covered in margin by category.

The feed is the common denominator

For an online shop the feed carries the volume, so the feed is what holds the expansion together. For Shopping ads Google lets you target several countries from a single offer and will convert currency automatically if you cannot supply prices in a foreign one. It does need to know where and at what cost you deliver, so the shipping attributes have to be right for every target country.

And one rule admits no exceptions: the currency and the language on the landing page have to match the target country. Google has its own requirements here and allows several combinations of language and currency, but none of them covers a page that shows an Austrian visitor prices in Czech crowns. The most common mistake in a fast expansion looks exactly like that: the ads are in German and the basket still counts in crowns.

Watch out for one quiet moment: the shop moving to a different platform. The feed URL changes, and often the format and the item identifiers too. If you have thirteen markets hanging off one feed, thirteen break at once — and because the campaigns don't stop, they just run out of things to show, you will notice it in a drop in revenue a day later than in an error message.

How to launch market number six

Once the template exists, launching another market stops being a project and becomes a list:

1. Feed. Add the country, fill in shipping and availability, check the currency and the landing page language.
2. Measurement. Connect orders from the new country to the same data source as the other markets. Without it you steer blind for the first month.
3. Campaigns from the template. Same set, same names, same labels.
4. Keywords and copy in the market's language. This is the one step that cannot be copied. It is worth paying someone who speaks it.
5. A starting target. Begin on the loose side and tighten once conversions accumulate. Too strict a target on a new market means the campaign barely gets going and has nothing to learn from.
6. A review date. Written down in advance: when you will look at the market and what has to be true for it to continue.

The last point is the one people skip most often, and it is the cheapest. Without it a new market sits in the account for years simply because nobody ever judged it.

Comparing markets side by side

This is the whole reward for the tidiness. When markets share a skeleton and the same metric definitions, you can build a single table: a row per market, cost and revenue in one currency, cost of sale, change against last month.

Two things such a report rests on:

• One currency. Convert to a single one, or you are comparing numbers that have nothing to do with each other. And hold one rate per period rather than a daily one; otherwise exchange-rate swings end up inside your campaign results.
• One source of orders. Ideally real orders from the shop, not the conversions ad platforms credit to themselves. Those overlap between platforms, and across thirteen markets the difference adds up into a number you cannot decide on.

That report is also the only honest basis for the question "where do we add budget". The market with the lowest cost of sale is not automatically where the next money belongs — it may simply be small and already saturated. What you want to know is where there is still demand you are not buying.

What to take from this

A shared structure across markets is not tidying for its own sake. It is the condition for being able to say which market earns at all. Keep the same what gets named and measured: the set of campaigns, the names, the labels and the metric definitions. Keep different what the market dictates: budget, target, prices, shipping and language.

And once you have more than a couple of accounts, write changes down — who, what, why, and when it gets judged. Without that you cannot work out six months later what caused the drop on the fourth market. We built ClientDeck for this, but a spreadsheet is infinitely better than nothing.

If you are dealing with this across several markets and don't want to build it yourself, this is what campaign management at Independency looks like.