Last click or data-driven for a food and beverage store?

If you sell food and drink, the two models usually agree on quick reorders, because a one-visit order has nothing to split. Data-driven earns its place on first orders that take several visits. Neither model can credit tastings, shop shelves or renewals charged without a visit, so judge those another way.

If you sell food and drink

If you sell coffee, tea, snacks, sauces or drinks, your buyers often decide fast and come back. A reorder tends to be one visit: an email click, a bookmark, a typed address. The first order is a different animal. It may follow a tasting, a shelf, a video and two visits to your site.

That split decides the model question for you. Last click and data-driven only disagree when a path holds more than one channel. Quick reorders rarely do. First orders more often do, and that is where data-driven can hand some credit to the channel that found the buyer.

Subscriptions add a twist. If renewals are charged automatically, there is often no visit before them. A browser tag may never record those orders, and no model can credit a touch that never happened.

Here is what one store's export shows. It is one store, and nothing in the export says it sells food or drink.

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.

In this one store, 1-touch journeys hold 79.5% of revenue on the Journeys sheet, with 0.5 days to buy. Every model gives those sales to the same single touch. If you sell food and drink, a quick reorder looks exactly like that row.

Journeys with 2 to 3 touches hold 12.2% of revenue on the Journeys sheet and took 12.5 days to buy. That slower shape is closer to a considered first order, such as a gift box or a first subscription. It is where the two models can part ways.

Direct holds 57.7% of revenue on the Channels sheet in every view. The export cannot show how much of it came from reorders and how much from first orders. That split is the one to find in your own data before you pick a model.

What changes for a food and beverage store?

Reorders blur the comparison. GA4's models skip direct visits unless a whole path is direct, and any touchpoint inside the lookback window stays eligible for credit. So a reorder typed straight into the browser can be credited to an older click. GA4's last click can then pay an old ad for a habit your product built.

Shopify draws the line elsewhere. Its marketing reports treat the first referrer after an order as a first interaction for the next order. So Shopify starts each reorder fresh. Expect Shopify and GA4 to disagree most about your most loyal buyers.

Seasons change the mix. Gift boxes in December can bring buyers who take several visits, while summer drinks may sell in one. Compare the models within the same season, year on year, not December against March.

Shelves and tastings stay invisible. If you also sell in shops or at markets, a lot of discovery can happen offline. Neither model can weigh it, so answer that question with a regional test, not a model setting.

What to do this week

1. Split new and returning buyers. In Shopify admin, go to Analytics > Reports, filter the Category to Customers and open New vs returning customers. Pass: you know what share of last quarter's customers came back. Fail: you were about to judge both models on a blend of first orders and reorders.
2. Size your subscriptions. In the Sales reports, open Subscription vs one-time sales, which fills in if you use the Shopify Subscriptions app. Pass: subscriptions are a small slice, so most revenue has a visit a model can credit. Fail: they are a big slice, so check that renewal orders reach GA4 before you trust any model.
3. Compare the models in two seasons. In GA4, open Advertising > Attribution > Attribution models and pick purchase. Set Data-driven against Paid and organic last click for your last gift season, then for a quiet month. Pass: channels keep similar credit in both. Fail: credit swings in the gift season, so judge seasonal channels on that season's numbers.

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: Get started with attribution (Google); Select attribution settings (Google); Key event attribution models report (Google); Marketing reports (Shopify); Customers reports (Shopify); Sales reports (Shopify)

Related answers

• All Attribution models questions
• Last click vs data-driven attribution, which is better?
• How to test data-driven against last click before switching
• Repeat purchase rate, defined