How do I test branded search for a food and drink store?

Usually like any brand test, with two changes if you sell food and drink. Split first orders from reorders, because regulars may search your name just to buy again. And check whether retailers that stock you bid on your name. Then pause brand ads in some regions and compare total sales with the rest.

If you sell food and drink

If you sell coffee, snacks, sauces or drinks, the products run out and good customers come back. Some come back by typing your name into Google, because that is quicker than digging out an old order email. A brand ad above your own listing then collects a sale that was settled before the search. Reorders like these can make brand ads look like the best line in the account.

If supermarkets, delis or online grocers also stock you, some of them may bid on your name. Google's trademark policy does not restrict trademarks used as keywords, so nothing stops them. People also search a food brand for recipes, stockists and opening hours. Search Console counts all of those as branded queries if they include your name.

One store's export, read for repeat buying

One store's anonymised GA4 export, 1 January 2024 to 21 August 2026. It holds shares of revenue only: no ad spend, no order counts. Nothing in the file says Store A sells food, so treat these rows as one pattern to hold against your own.

On the Channels sheet, Direct takes 57.7% of that store's revenue in all three views. Direct, in GA4's terms, means a bookmark or an address typed in by hand. If your regulars reorder that way, a paused brand ad costs you nothing with them, because they never see it.

On the Journeys sheet, one-touch journeys hold 79.5% of revenue and take 0.5 days to buy. That is what settled buying looks like in an export. If your reorders arrive in one visit, they sit in this row, whichever channel brought that visit.

Journeys of 2 or more touches add up to 20.6% of that store's revenue, about a fifth. If your first-time buyers take a few looks before they buy, they sit in this smaller pool. It is also where a brand ad has the best chance of changing a mind.

The export cannot tell a reorder from a first order, and it holds no spend. So it cannot say what brand ads did for that store. It is one store, not a benchmark, and the split a food brand needs lives in Shopify.

What changes for a food brand's brand test?

• Score first orders apart from reorders. Shopify counts a customer as returning once their order history already holds an order. Read the test on both lines. If first orders hold steady where the ads stopped, the brand campaign was mostly charging you for reorders.
• Keep subscriptions out of the totals. If you sell subscriptions, renewals arrive without any search at all. Shopify reports subscription sales separately, so take them out of both groups before you compare.
• Watch where a lost click goes. If a stockist's ad takes the top spot while yours is paused, the buyer may still buy your product, just from them. Your Shopify sales dip, but the product still sold. Judge that gap in margin, not in sales.
• Pick a calm stretch. Food sells to a calendar: gifting before the holidays, cold drinks in summer. Run the test when both groups face the same weeks, and finish before your peak so the result can set peak spending.

What to do this week

1. Split recent sales by customer type. In Shopify admin, go to Analytics, then Reports, click the Category filter, choose Customers and open New vs returning customers. Pass: first-time sales move steadily from week to week, so they make a usable baseline. Fail: they jump with every promotion, so pick test weeks with no promotion at all.
2. Look for stockists in your brand auction. In Google Ads, go to Campaigns, tick the brand campaign and select Auction insights. Pass: no retailer or marketplace appears often next to your ad. Fail: one does, so expect your brand ads to defend real sales and test in fewer regions.
3. Separate buying searches from recipe searches. In Search Console, open Performance, click the filter bar, choose Query and pick Custom (regex), then Doesn't match regex with words like recipe. Pass: plain brand searches form their own weekly line. Fail: the line is too thin to read weekly, so read it monthly and plan a longer test.

Check the homework. Your GA4 Attribution paths export already holds the evidence. Causality Engine reads that one file and shows what each channel caused next to what last-click gave it, in 1 to 2 minutes, for €99 once (excluding VAT), refundable within 30 days. Check the homework

Sources, 1 October 2026: Trademarks (Google Advertising Policies Help); Performance report (Search results): Common tasks and use cases (Search Console Help); Default channel group (Google Analytics Help); Customers reports (Shopify Help Center); Sales reports (Shopify Help Center); Use auction insights to compare performance (Google Ads Help); Performance report (Search results): Advanced filtering and comparison (Search Console Help)

Related answers

• All Incrementality questions
• How do I test branded search incrementality?
• How to test branded search incrementality, step by step
• How should a food and beverage store split its ad budget?
• Control group, in the glossary