Marketing Attribution for Ecommerce: Models, Tools and What to Trust

Key Takeaways

• Marketing attribution decides which touchpoints, such as ads, searches and emails, get credit for a sale.
• Rule based models share credit by a fixed rule; data driven attribution uses an algorithm based on real paths.
• Google removed first click, linear, time decay and position based models from Google Ads and GA4 in 2023, leaving data driven and last click.
• Ad platforms each claim credit for any sale they touched, so their reported revenue usually adds up to more than the store took.
• Blended measures, such as total revenue divided by total ad spend, are the check on every platform's claims.
• For big decisions, test incrementality; Google made its open source marketing mix model, Meridian, available to everyone in January 2025.

Quick Answer

Marketing attribution is how an online store decides which ads, searches and emails deserve credit for each sale. Models range from simple rules, such as first click or last click, to data driven attribution. No model is exact, and each ad platform credits itself generously, so the practical approach for a Shopify store is to fix tracking, use consistent UTM tags, compare channels with one model in Shopify or GA4, check blended revenue against total ad spend every week, and test incrementality before big budget changes.

Table of Contents

• What Is Marketing Attribution?
• Why Attribution Is Hard for Online Stores
• Attribution Models Explained
• How Google, Meta and Shopify Attribute Sales
• Blended Measures: The Check on Every Platform
• Beyond Attribution: Incrementality and Marketing Mix Modelling
• A Practical Attribution Setup for Shopify Stores
• Why Matters Perspective
• Frequently Asked Questions
• Final Thoughts
• Related Articles

A typical online customer sees an Instagram ad, searches the brand on Google a few days later, signs up for the newsletter and buys from an email. Which channel earned the sale? Every platform involved will say it did. Attribution is how you decide, and getting it roughly right is what lets you spend money with confidence.

What Is Marketing Attribution?

Marketing attribution assigns credit for a conversion, usually a purchase, to the marketing touchpoints that led to it. Single touch attribution gives all the credit to one touchpoint; multi touch attribution shares it between several. The point is practical: to see which channels and campaigns are working so you can move budget to them.

Why Attribution Is Hard for Online Stores

• Many touchpoints: customers often meet a brand several times across channels before buying.
• Every platform marks its own homework: Google, Meta and TikTok each count any sale they touched.
• Gaps in tracking: ad blockers, privacy settings, declined cookies and switching devices hide parts of the journey.
• Views versus clicks: some platforms count people who saw an ad without clicking.
• Broken setups: broken conversion tracking is the most common problem we find in inherited Google Ads accounts this year; in the worst case every purchase was counted twice, doubling the reported ROAS.

Attribution Models Explained

In words: for a journey of a Meta ad, then a Google Shopping click, then an email, first click gives Meta all the credit, last click gives the email all of it, linear gives each a third, and position based gives 40% each to Meta and the email and 20% to Google Shopping. Same sale, four different answers.

Google has narrowed the choice in its own tools. It announced that first click, linear, time decay and position based models would be removed from Google Ads and Google Analytics 4, with data driven attribution, last click and external attribution unaffected; in Google Ads, conversion actions still using the removed models moved to data driven attribution from September 2023.

How Google, Meta and Shopify Attribute Sales

Shopify's marketing reports, introduced in February 2023, let you switch between first click, last click and last non direct click, which makes them a useful neutral view alongside the ad platforms.

Blended Measures: The Check on Every Platform

Because each platform credits itself, add up what they claim and compare it with what the store actually took. A worked example for one month:

The platforms are not lying; they each counted sales they genuinely touched. But their totals cannot all be right, so use the blended figure to judge overall efficiency and the platform figures to compare campaigns within each platform. Count new customers separately: one client's Performance Max campaign was spending 65% of its budget on brand searches before it came to us, and after brand exclusions first time customers rose 52% the following month.

Beyond Attribution: Incrementality and Marketing Mix Modelling

Attribution shares credit for sales that happened. It cannot tell you whether a sale would have happened anyway. Two methods answer that:

• Incrementality tests: compare a group that saw the ads with a similar group that did not, for example by pausing a channel in some regions or using a platform's holdout test.
• Marketing mix modelling: statistical models of sales and spend over time. Google made Meridian, its open source marketing mix model, available to all marketers on 29 January 2025; it can measure sales or profit and include factors such as pricing and promotions.

Both need enough spend and history to produce reliable answers, so they suit established brands spending across several channels. Our Triple Whale review covers attribution tools that bundle these methods.

A Practical Attribution Setup for Shopify Stores

1. Fix tracking first: one pixel per platform, purchases recorded once, and consent set up properly. See Meta Pixel and Conversions API.
2. Tag everything with UTM parameters using one naming convention for source, medium and campaign.
3. Choose one model for channel comparison, such as last non direct click in Shopify, and stick with it.
4. Take each number from its source: revenue from Shopify and spend from each ad platform, so a broken connection cannot mislead you.
5. Report new customers by channel, not just revenue.
6. Check the blended return weekly, against a target set from your margin; our ROAS calculator shows break even.
7. Test before big changes: pause or cut a channel in one region for a few weeks and watch total sales.
8. Think in lifetime value: a channel can look weak on first orders and still be profitable; one coffee client's TikTok Shop campaign had a negative ROAS but brought first time customers whose lifetime value made it profitable.

Why Matters Perspective

We do not believe any attribution model tells the whole truth, and we are wary of reports that show every channel performing brilliantly at once. What works is a small set of measures everyone trusts: correct tracking, new customers by channel, the blended return against a margin based target, and occasional tests to check the big assumptions.

In our own client reports we take each KPI from as close to its original source as possible: revenue from Shopify, ad spend from each ad platform directly. That is about credibility. We cannot afford to be wrong with the data, and every tool or connection in between is one more thing that can break or define a KPI differently.

We work to a floor of at least 2x ROAS across the accounts we manage, and ad accounts we set up always belong to the client. We are a Shopify Select partner and a Klaviyo, Meta and TikTok partner. Our pricing is published, PPC management starts from £650 a month plus VAT, and clients are billed one month in arrears with no long contracts. For a free review of your tracking and attribution, visit our PPC agency page.

Frequently Asked Questions

What is marketing attribution?

Marketing attribution is the practice of deciding which marketing touchpoints, such as an ad, a search result or an email, deserve credit for a sale. It helps an online store see which channels are working, so budget can be moved towards them.

What are the main attribution models?

Rule based models include first click, last click, last non direct click, linear, time decay and position based, each sharing credit by a fixed rule. Data driven attribution uses an algorithm to share credit based on the paths that did and did not lead to conversions. Google Ads now offers data driven and last click only.

Which attribution model is best for ecommerce?

No single model is right. Use data driven or last click inside each ad platform for bidding, a consistent model such as last non direct click in Shopify's reports for comparing channels, and blended figures such as total revenue divided by total ad spend to check the whole picture.

Why do Google, Meta and Shopify report different numbers?

Each counts conversions with its own rules, time windows and data. Ad platforms credit themselves for any sale they touched, so their reported revenue often adds up to more than the store took, while Shopify only sees what its own tracking and UTM tags record.

What is the difference between attribution and incrementality?

Attribution shares credit for sales that happened between the touchpoints involved. Incrementality asks whether the sale would have happened anyway, usually by comparing a group that saw the ads with a group that did not. Attribution guides day to day decisions; incrementality tests check the big ones.

What is marketing mix modelling?

Marketing mix modelling uses statistics on sales and spend over time to estimate how much each channel and other factors, such as pricing and promotions, contribute. Google released Meridian, an open source marketing mix model, to all marketers in January 2025. It suits brands with large budgets across several channels.

How should a small Shopify store measure attribution?

Fix tracking first, tag every ad and email with consistent UTM parameters, compare channels in Shopify's marketing reports, and watch total revenue against total ad spend each week. A paid attribution tool is worth adding once spend across several channels is large enough for better data to change decisions.

Final Thoughts

Marketing attribution will never be perfect, but it does not need to be. A Shopify store that tracks correctly, compares channels with one consistent model, checks the blended return each week and tests its big assumptions will make better budget decisions than one chasing the perfect model.

Start with tracking and UTM tags, add a paid tool when your spend justifies it, and keep lifetime value in view. If you would like us to review your setup, email us.

Related Articles

• Triple Whale Review
• Meta Pixel and Conversions API on Shopify
• What Is a Good ROAS for Ecommerce?
• ROAS Calculator
• Customer Acquisition for Ecommerce
• Google Ads Audit Checklist for Ecommerce
• Ecommerce Marketing Strategy