Which attribution tools match Shopify orders to Meta and Google ads?

Order-based attribution apps are the tools that connect directly to Shopify and match real orders against Meta and Google ad clicks. They read every order from your store, then work backwards to the ad clicks and visits that came before it, using click IDs (gclid, fbclid) and UTM tags. Triple Whale, Northbeam, Polar Analytics and Prism all work this way. Shopify's own marketing reports do a lighter version for free.

Two other kinds of tool get lumped in with them. Meta Ads Manager and Google Ads report on their own clicks and views, so every platform counts the sales it thinks it caused. Server-side tracking apps like Elevar and Littledata send your orders to Meta and Google, so the platforms see more of your sales. That helps the platforms, but it doesn't tell you which channel earned the order.

Knowing which of these you need is most of the buying decision. The rest of this post covers what each kind can see, how one order gets matched to an ad, and the questions to ask before you pay.

The four places your numbers come from

What it can do	Ad platform reports	Shopify's own reports	Order-based attribution apps	Server-side tracking apps	Prism
Starts from your actual Shopify orders	–	✓	✓	Uses the order, to send it on	✓
Sees the whole journey across channels	–	✓	✓	Only by feeding GA4	✓
Each order is credited once, not once per platform	–	✓	✓	–	✓
Matches on ad click IDs, not just UTMs	Its own clicks only	Mostly UTMs and referrers	✓	✓	✓
Shows spend and ROAS per channel and campaign	Its own channel only	Only where an ad app reports cost	✓	–	✓
Sends missed sales back to Meta and Google	–	Via the Meta and Google apps	Some do	✓	Google and Meta

1. Ad platform reports (Meta Ads Manager, Google Ads)

Each platform counts purchases it can tie to its own ads. Meta's default counts a purchase within 7 days of a click or 1 day after someone saw the ad. Google Ads uses data-driven attribution with a 30-day click window by default, and its Conversions column mixes observed and modelled conversions.

Those rules are reasonable on their own. The trouble is that each platform only sees itself, so when Meta and Google both touched an order, both count all of it. Add up what every platform claims and you'll usually get more revenue than Shopify recorded. We went through why in detail for Meta and Google.

2. Shopify's own marketing reports

Shopify already matches orders to the visits before them. Its marketing reports use last non-direct click by default, and you can switch to last click, first click, linear or any click (Shopify Help). It's free and it's built on your real orders, which already puts it ahead of reading Ads Manager alone.

The limits are worth knowing:

• It mostly reads UTM tags and referrers. An untagged ad click often shows up as direct.
• Spend, ROAS and CAC only appear for channels whose app reports cost to Shopify. Everywhere else you get a dash.

Shopify also changes what these reports include from time to time. If your channel totals jump overnight, check Shopify's changelog before you change your ad spend.

3. Order-based attribution apps

An order-based app does three things:

1. It records visits with its own tracking script, saving the click IDs and UTM tags on each landing page.
2. It reads every order from Shopify, not just the ones its script happened to see at checkout.
3. It links each order back to the visits before it and splits the credit using an attribution model. One order, credited once, across every channel that touched it.

Then it puts your ad spend next to that credit, so ROAS is calculated on revenue your store actually took. The tools differ mostly in price and in what else they bundle:

• Triple Whale has a free plan with first and last click. Paid plans add multi-touch attribution and are priced by your store's annual GMV (pricing guide).
• Northbeam starts at about $1,500 a month, quoted per brand, and is built for large ad budgets (pricing).
• Polar Analytics is priced by online GMV, from about $749 a month on its Shopify listing (App Store).
• Prism starts at $29 a month for up to 1,000 orders, priced by order count. More on how it works below.

We've compared Prism side by side with Triple Whale, Northbeam and Polar Analytics, with prices at 1,000, 5,000 and 10,000 orders a month.

4. Server-side tracking apps

These apps exist because the browser pixel misses sales. Safari limits, ad blockers, consent banners and off-site checkouts all stop it from firing. A server-side app takes the order from your store and sends it to Meta's Conversions API, Google Ads and others directly, with no browser involved.

• Elevar is priced by orders, about $225 a month for 2,000 (App Store). Elevar is now part of Audiense.
• Littledata starts at $0.35 per order with no minimum, or $199 a month for 1,500 orders ($159 if billed annually) (pricing).

How they line up against Prism: Prism vs Elevar, Prism vs Littledata, and every comparison.

That makes the platforms' numbers more complete, which helps their bidding. It still doesn't tell you which channel deserves the order, because the platform still only credits itself. If that gap is the thing bothering you, why ad platforms miss your Shopify sales goes deeper.

How matching an order to an ad actually works

"Matches orders to ad exposure" covers four different methods. Good tools use more than one.

Click IDs. When someone clicks a Google ad, Google adds a gclid to the link. Meta adds fbclid, TikTok ttclid, Microsoft msclkid. If the tracking script saves that ID on the visit, and the order is linked to that visit, you know which platform sent the buyer. It's the strongest signal available.

UTM tags. The utm_source, utm_medium and utm_campaign values you add to your links. They're only as good as your tagging. An ad with no UTMs, or a link shortener that strips them, quietly turns paid traffic into direct. Our UTM setup guide has the exact values for each platform.

Referrers. The site the visitor came from, like instagram.com or google.com. Enough to tell organic social from organic search, but not a paid click from an unpaid one.

Modelled or survey data. Some tools add media mix modelling, or a post-purchase "how did you hear about us?" survey. Both are useful for channels that don't leave a click, like podcasts or TV. Neither is a match to a specific ad click.

Here's one order walked through:

1. Day 1: a shopper taps your Meta ad. The link carries fbclid and your UTMs. The tracking script saves both against an anonymous visitor ID.
2. Day 4: they come back from a Google search ad. gclid gets saved on that visit.
3. Day 6: they type your URL in and buy. Shopify sends the order to the attribution tool.
4. The tool finds the visitor's three visits in the last 30 days, sees one Meta paid click, one Google paid click and one direct visit, and splits the order between them using your chosen model.

Every model gives out exactly 100% of the order between those three visits. How the six models compare on one order shows what each split looks like.

The orders the tracking script never saw

Tools differ most here. Some orders arrive with no recorded journey: the buyer used a one-click checkout, cleared their cookies, or blocked the script. A tool that only credits orders its own script saw at checkout drops those orders. A tool built on the order still records them, and either recovers where they came from or shows them honestly as unattributed.

Ask every vendor what happens to these orders, and what share of your orders it expects them to be.

9 questions to ask before you buy

1. Does it read every order from Shopify, or only the ones its script saw at checkout?
2. Does it match on click IDs as well as UTMs? UTMs alone mean every untagged click becomes direct.
3. Can I open one order and see its full journey? If you can't check a single order, you can't trust the totals.
4. What happens to orders with no tracked visits? Dropped, guessed, or shown separately?
5. Does it handle cash on delivery and external checkouts? If you sell COD or use a one-click checkout like GoKwik, ask specifically.
6. How is it priced: orders, GMV, or pageviews? GMV pricing goes up when your prices do, even if your order count doesn't.
7. Do its attributed orders add up to your Shopify order count? Every order should be credited exactly once. If the tool's total is higher or lower than Shopify's for the same dates, ask why.
8. Does it send missed sales back to Meta and Google? Not essential, but it saves a second app.
9. Is there a free trial on your own store's data? A demo account tells you nothing about your tagging.

Where Prism fits

Prism is an order-based attribution app for Shopify. It reads every order from Shopify, matches it to the visits before it using click IDs and UTMs, and splits the credit across six attribution models you can switch between. For orders its tracking script missed, it tries to recover the source from Shopify's own record of the customer's journey and the order's checkout details before marking the order as unknown.

It can also send sales the browser pixel missed back to Google Ads and Meta through their server-side APIs, so you don't need a separate tracking app for that. Plans start at $29 a month for up to 1,000 orders, with a 14-day free trial.

Prism also has a separate profit and loss view. It is its own feature. Attribution only decides which channel earned an order, and never changes your total revenue or profit.

Quick answers

What's the best attribution tool for a small Shopify store? Start with Shopify's own marketing reports and clean UTM tags. When you need ad-level detail, spend and ROAS in one place, or multi-touch credit, move to an order-based app priced by order count.

Is Shopify's built-in attribution good enough? For a store with one or two channels and tidy UTMs, often yes. It gets harder once you run several paid channels, need ad-level ROAS, or have a lot of traffic showing up as direct.

Do I need both an attribution app and a server-side tracking app? Only if your attribution app doesn't send conversions back to the platforms. They do different jobs: one decides credit, the other makes Meta and Google's own numbers more complete.

Why do Meta and Google together claim more sales than I made? Each one credits itself for any order it touched, so an order both platforms touched gets counted twice. An order-based tool counts it once and splits it.