Google Ads for an online store: what to run before Shopping arrives.

If you sell online in Bulgaria, Google gives you fewer tools today than it gives a store in Romania or Greece: Shopping ads and free product listings have not launched here yet. That is not a reason to wait. It is a reason to set up what already works properly, so that when Shopping does arrive your account has data behind it instead of starting from zero.

What you can actually run today

Without Shopping, the main tool for an online store on Google is still Search - the text ads above the results. They reach people who are already looking for something specific, which is the shortest route to a sale. In practice the accounts that work are built from three pieces:

Search by category, not for the whole store

The most common mistake is one campaign "for the whole store" on broad keywords. The budget then flows to wherever clicks are cheapest, not to wherever the sales are. The better order is to start from the categories that carry revenue and give each its own budget.

• One category, one ad group. Mattresses, pillows and bed frames are not searched the same way and do not earn the same margin. Lumped together, you cannot see which one is actually working.
• The ad lands on the category. Someone who searched for an orthopaedic mattress should arrive among the mattresses, not on the home page hunting for them.
• The best seller gets the biggest share. Not an even split - budget goes where you already know there is demand and margin.

Keep the brand separate from everything else

People searching for your store by name have already decided. Any campaign that "catches" them will look brilliant in the report without having produced a single new sale.

This is where Performance Max most often misleads. The algorithm goes after the easiest conversions available, and the easiest are branded searches. The result is a high ROAS that is largely paid-for traffic which would have arrived for free.

The fix has two parts:

1. A small, separate Search campaign for the brand only. Low cost per click and its own budget, so it does not compete with anything else.
2. Brand excluded from Performance Max. Google has a setting for exactly this: in the campaign settings, under additional settings, you apply a brand list under brand exclusions. According to Google it covers Search, Shopping and YouTube search inventory - not Display, Gmail or Discover.

A practical threshold from an account audit published in June 2026: if branded searches make up more than 15% of a campaign's conversion value, you have a cannibalisation problem worth fixing.

How to split the budget

There is no universally correct split, but there is a logic to follow. This is how it looks for one of our clients, a retailer with several physical stores:

The logic behind the numbers: the main budget goes where demand is already proven; the brand gets little, because it is cheap and only needs guarding; Performance Max gets enough to collect data, but not so much that it crowds out Search.

How much to spend in total, and what management costs, is a separate question - we broke it down in how much Google and Facebook ads cost. The short version: below roughly 300 EUR a month in ad spend, the honest advice is not to pay an agency at all.

Data first, then optimisation

A new account does not know who buys. Tell it to chase revenue on day one and it has nothing to chase it with. So we move through three stages:

How many is enough? Google gives a minimum of 15 conversions in the past 30 days to use target ROAS on Search and Shopping, requires those conversions to carry values, and recommends waiting one to two conversion cycles before setting the target. Below that, the strategy simply has no data to bid on.

What changes when Shopping reaches Bulgaria

Shopping ads show the product with its picture and price right in the results. Where they exist, the click is usually cheaper than Search: on US e-commerce data compiled by Store Growers from WordStream, the average cost per click is 0.66 USD on Shopping against 1.16 USD on Search. Bulgarian numbers will differ, but the direction of that gap is why every store wants to be ready.

When exactly? Google has not announced a date. Several Bulgarian agencies expect a launch around October 2026, after the euro changeover, but that is an estimate, not a confirmation - Google's official Merchant Center changelog does not mention Bulgaria yet. We track it on a separate page we keep updated.

What to do in the meantime is clear. Get Merchant Center and the product feed ready now, because that takes weeks, not days - the feed, step by step. And when Shopping does go live, start with Standard Shopping rather than Performance Max - here is why.