Analytics and attribution
What is attribution in marketing?
Attribution is how a sale is credited to the ads, emails and channels a customer interacted with before buying. Each tool does it differently.
Attribution assigns credit for a conversion to the marketing touchpoints that came before it. A customer may see a TikTok ad, click a Meta ad, read an email and finally search for your brand on Google. Attribution decides which of these gets the credit, and how much. The rules come from an attribution model, such as last click or data-driven, and from an attribution window.
Every platform attributes conversions to itself using its own data and rules. Meta, Google, TikTok and Klaviyo can each claim the same order, so their reports added together often exceed your real sales. That is why attribution numbers are best used to compare campaigns within one platform. For budget decisions across channels, check them against total sales, MER and, where possible, incrementality tests.
E-commerce example
A customer sees your TikTok ad, clicks a Meta ad later that day, then searches for your brand on Google and buys for 80 EUR. Google Ads may count the purchase from the search click. Meta may count it within its click window. TikTok may count it as a view-through conversion. One 80 EUR order can appear as 240 EUR across three dashboards, while Shopify records 80 EUR.
Typical mistakes
- Adding up conversions from all platforms. The sum double counts and makes paid media look stronger than it is.
- Cutting top-of-funnel campaigns because last-click reports show no sales. Those campaigns often start the journeys that other channels close.
- Comparing platforms that use different windows and models as if they measured the same thing. Check the settings before you compare.
How we use it at Pleroma
We use platform attribution to choose between campaigns inside one platform, and store revenue with MER to set budgets across channels. When the two disagree, store revenue wins.
Questions and answers
Which attribution model should a store use?
There is no perfect model. Use the platform's default for day-to-day optimisation, keep it consistent, and judge the overall budget against total revenue and MER. Run holdout or geo tests when a big budget decision depends on the answer.
Why does Shopify show fewer sales from ads than Meta does?
Shopify credits orders based on tracked visits and UTM parameters, and it cannot see who only viewed an ad. Meta counts orders after clicks and views within its own window. Neither is wrong; they answer different questions.
Want us to look at these numbers in your store?
30 minutes on your store and channels. You hear what to fix first.
Book a call