How do I choose ad tracking software for my store?
What decision should the tool help you make?
Most stores shop for features. Shop for a decision instead, because it tells you which features you can ignore.
- "Which ads should I turn off or scale?" You need ad-level sales matched to orders, refreshed often enough to act on.
- "How should I split budget between Meta and Google?" You need both channels judged on the same orders, with no sale counted twice.
- "What does a new customer cost me?" You need new and returning buyers split per channel.
- "Is TV or a podcast working?" Click tracking will not answer this. You need surveys, holdout tests or marketing mix modeling.
The 7 questions to ask every vendor
- Does every sale in the tool map to a real order? Ask to click from a channel's sales number down to the order list behind it. If you cannot, part of the number is modeled.
- Do the channel totals add up to my store's sales? Some models are built to exceed them. Shopify itself says its "any click" model "allocates more credit than orders you've received" (Shopify Help Center). That is fine for analysis, but not for budget decisions.
- Does it show the platform's own number next to its own? The gap is the point. In our sample Meta reported more than we matched in 68% of stores, and Google Ads reported less than we matched in most of 65 stores (median 0.81x).
- Does it send verified purchases back to Meta and Google? If it does, ask how it avoids double counting. Meta counts a browser and a server purchase as one only when both carry the same event ID (Meta for Developers).
- Does it support my store platform and every paid channel I run? WooCommerce, TikTok and Pinterest support varies a lot between tools.
- How long until the numbers are usable? Northbeam quotes 2 to 4 weeks for Shopify brands and 4 to 8 weeks for others (Northbeam docs). Triple Whale's help center says its pixel only sees visitors from the day it was installed, so wait the length of your average customer journey before acting (Triple Whale, archived November 2025).
- What will it cost now, at twice my size, and how do I leave? Get the price steps and the notice period in writing. What a good trial and contract include is in what an attribution tool's trial and contract should include.
Which questions matter most for your store
| Your situation | Weigh most | Can skip |
|---|---|---|
| One channel, under about $10k a month in sales | Question 4 (data back to Meta) | Multi-touch models, MMM |
| Meta and Google together | Questions 1, 2 and 3 | Creative dashboards, until the totals add up |
| WooCommerce store | Questions 1 and 5 | Anything built only for Shopify |
| Agency with many client stores | Questions 3 and 7 (per-store price) | Features only one client needs |
My opinion: questions 1 and 2 decide whether you can move money on the tool at all. A tool that fails them is a nicer window onto the platforms' own numbers, which you already have for free.
How to test the finalists
- Install your two finalists next to each other. Change nothing in your ad accounts.
- Wait 14 days, or longer if most of your buyers take more than a week to decide.
- For the same dates, compare each tool's total sales with the orders in your store.
- Pick 5 orders and check that each tool names the same ad for them.
- Keep the one that adds up, and that you actually opened during the test.
Do not judge on the demo. A demo runs on a sample store with clean data, and your store will not look like that. How to run two tools side by side without losing data is in how to switch attribution tools safely, and the scoring version of these questions is in the guide how to choose an attribution tool. TrueROAS was built to answer yes to these questions, so hold us to the same list as everyone else.

