How to test your Shopify purchase tracking
By the HonestTag team · Published November 6, 2026 · Updated November 11, 2026
A test order tells you one thing well. Real orders tell you the rest.
A test order shows whether your purchase event fires and reaches the platform, with the right order value. It cannot show that an ad will get credit, because a test order has no ad click behind it, so a Google Ads conversion column that stays at 0 after your test is the expected result. Check the test where each platform shows events as they arrive. Then check one normal week of real orders: count what each platform received against what Shopify took. That week, not the test, is what tells you your tracking works.
What a test order can prove, and what it cannot
A test order can show three things: the purchase event fired, it carried the right order number and value, and the platform received it. That is worth knowing before you spend money on ads.
What it cannot show is attribution. Google describes how it records a conversion: when someone clicks your ad, "a temporary cookie is placed on their computer or mobile device. When they complete the action you defined, our system recognizes the cookie (through the code snippet you added), and we record a conversion." If you open your store by typing its address and buy, there is no ad click for Google to recognize, so nothing lands in the Conversions column. Your tag can be working perfectly while that column reads 0. The reverse also applies: if your browser clicked one of your own ads recently, that click can still be on record, so use a private browsing window for the test.
Two more things a test order hides. A new Google Ads conversion action often shows a status other than Active at first, and Google says "An "Unverified" or "Inactive" status is normal for new tags." It changes to Active "only after Google Ads records its first conversion or a successful test event." And your own visit is not a typical shopper's. If your store asks for consent, Shopify's pixel testing guide says a pixel may not load until you have "provided consent through your cookie banner." You probably clicked accept. Some of your shoppers will not, which is one reason real counts come in lower than orders. Cookie banners and conversion counts covers that side.
Where a test purchase shows up
Each platform has a place that shows events as they arrive. That is where a test purchase belongs, not in the ad reports.
| Where to look | What it shows you |
|---|---|
| Shopify Pixel Helper | A custom pixel's events on your online store "in real-time, alongside the data for those events" |
| Meta Events Manager, Test events | Browser events from a window you open from that page. Meta says activity "is only recorded while the test events web page is open" |
| Google Tag Assistant | Whether the Google tag fires on your pages. Google points to it "for verification" |
| Google Analytics DebugView | Events Analytics collects from your device "in real time", once Tag Assistant or preview mode turns debug mode on |
Server-sent purchases are the exception that confuses people. Meta's Test events window shows a server event only when it is sent with the test ID the tool generates: "Send the test ID as a test_event_code parameter to start seeing event activity appear in the Test Events window." Meta says the test_event_code field "should be used only for testing," so a server purchase for a real order, sent without it, does not show in the Test Events window, even when it arrives. Those purchases show in Events Manager after real orders come in. Meta says data "may take 30 minutes or more to appear" after you set up an event.
Meta's tool also helps when two tools send the same purchase: it lets you "Check which events are processed and which are deduplicated." If your test order shows two Purchase events and neither is marked as a duplicate, read duplicate purchase events before you go further.
Test gateway or real order
Shopify gives you two ways to place a test order: its test payment gateway, or test mode if you use Shopify Payments. Neither charges you, and Shopify says "Test orders and simulated transactions don't display in your payouts or reports." Both need a paid Shopify plan.
Both also take your checkout out of live mode. Shopify says "Customers can't place live orders during the time that your payment providers are in test mode", and of Shopify Payments test mode, "you shouldn't activate test mode on a store that's in production." Part of the reason: "Customers might still be able to complete checkout through express payment options such as Apple Pay or Google Pay, but no real charges are processed." So a test order is for a store that is not selling yet, or for a development store; Shopify's pixel guide suggests testing pixels "on your dev store to ensure that they work as expected before adding them to your live online store." On a store that is taking orders, use a real order instead, described below. For a test order, Shopify warns "Don't fulfill any test orders, because you are charged for any shipping labels that you purchase", and "If you use an app that automatically fulfills orders, then deactivate it before you create test orders."
Shopify marks these orders as tests. Its order data says test orders "are made using the Shopify Bogus Gateway or a payment provider with test mode enabled", and that a test order "can't be converted into a real order and vice versa." Tools that read your orders can see that mark, and some skip test orders on purpose so they never report a sale that did not happen. So a test-gateway order is a fair test of the tags in a shopper's browser, but it may tell you nothing about a tool that sends purchases from a server.
The other way, which Shopify also offers, is a real order: "process an order using a real payment provider, and then immediately cancel and refund the order." You may pay processor fees, but your checkout stays live for everyone else. Expect it to be counted like any other purchase wherever it is sent. On Meta, even test activity counts: Meta says that after you clear your test activity it "will still appear in your Meta Events Manager, Ads Reporting, and Audience Creation results", and that server events sent with its test ID "are not dropped." Place the few test purchases you need and stop.
A worked example
The numbers below are made up to show the arithmetic. They are not from a store.
Before your store opens, you place a $42 order through Shopify's test gateway. Meta's Test events window shows a Purchase for $42 from your browser. The next day Google Ads still shows 0 conversions. Both results are right: the event left your store with the right value, and there was no ad click for Google to credit.
Once you are selling, you check a real week. Your store sends Meta each purchase twice, once from the browser pixel and once from a server. From Monday to Sunday Shopify shows 118 online store orders, so up to 236 copies can reach Meta. Events Manager's Total events lists 224 Purchase events, and Ads Manager credits 31 purchases to your ads.
- 224 of a possible 236 is your tracking check. Both senders reach Meta for most orders. Meta says "Events are not deduplicated in the Total events count", so about two per order is what two senders look like there, not a fault. Whether Meta pairs the copies is what the Test events tool shows. A few missing is expected, for example from shoppers who did not give consent.
- 354, three per order, would mean a sender you did not count.
- 31 is Meta's attribution: the purchases it credits to people shown your ads. It says nothing about whether the tag works. Events Manager vs Ads Manager explains why it sits so much lower. If those counts fall from one week to the next while your orders hold, a sudden drop in Meta purchases shows how to find what changed.
Check your tracking against your real orders. The free Mirror plan shows what Google Ads and Meta claim beside your own orders, and on paid plans every order shows what each platform was sent or why it was not.
What this means
A test and a real week answer different questions. A test purchase that shows in the platform's live view with the right value means the event fires. A Google Ads Conversions column at 0 after a test you reached by typing the address is expected and says nothing about the tag.
For the real week, take Shopify's online store orders for seven full days, times the number of tools sending purchases, and set that beside what the platform received for the same days in the same time zone; on Meta that is Total events in Events Manager. Close to that figure means your senders are reaching the platform. Clearly more points to a sender you did not count. Far fewer means events are going missing, often through consent or something that changed recently. Zero means the tag or connection is broken, or a consent setting is stopping every purchase event.
An uncounted sender inflates what the platform learns from, and missing events make ads that sold look as if they did not. HonestTag keeps the order side of this check: it never sends a Shopify test order to an ad platform, and on paid plans The Gap sets each platform's reported purchases beside the purchases HonestTag sent it.
How HonestTag helps
HonestTag cannot make a platform credit a test order. What it gives you is the order side of the check, kept for you.
- On the free Mirror plan, Live activity in Orders shows when your storefront pixel last sent an event, so you can confirm it is running. The Gap's "Who brought your customers" shows what Google Ads and Meta claim for the 14 days ending yesterday beside the new customers whose first recorded ad click or ad-tagged link was for that platform, from your own orders. It sends no purchases to any ad platform, and critical alerts when tracking breaks are emailed on every plan.
- On paid plans, Live activity shows page views from your storefront within a few seconds and a line for each order. HonestTag never sends a Shopify test order to an ad platform, and that order's line says so: "not sent: Shopify test order".
- On paid plans, The Gap sets "Platform-reported purchases" beside "Purchases HonestTag sent it" for each platform, so what a platform credits itself, like the 31 in the example, sits next to what it was sent. Purchases that were not sent are counted, with the reason.
- On paid plans, order proof on each order shows what each platform was sent or why it was not, so a missing purchase has an answer you can open.
- On paid plans, every verified web purchase is sent from HonestTag's server, within your plan's limit, so it does not depend on the shopper's browser. The event to look for is HonestTag_Purchase, which counts every purchase HonestTag sends, new and returning customers both. The server-side delivery guide covers the rest.
Install HonestTag free from the Shopify App Store and see, on the Mirror plan, what Google Ads and Meta claim beside the new customers your own orders show they first brought.
The free tracking check sets out how to read a normal week of the purchases each platform reports against your Shopify orders for the same days.
Frequently asked questions
Why doesn't my test order show as a conversion in Google Ads?
Google records a conversion when it recognizes the cookie left by an ad click. A test order you reach by typing your store's address has no ad click behind it, so 0 is the expected result. Use Tag Assistant to see the tag fire, and judge the Conversions column on real orders that followed ad clicks.
Why doesn't my server-side purchase show in Meta's Test events?
Meta's Test events window shows a server event only when it carries the test ID the tool generates. Meta says that ID is only for testing, so a server purchase for a real order, sent without it, does not show there, even when it arrives. Those purchases show in Events Manager after real orders arrive. Meta says data may take 30 minutes or more to appear after you set up an event.
Should I test with the test gateway or a real order?
On a store that is not selling yet, or a development store, use the test gateway or Shopify Payments test mode: no payment is charged and the order stays out of your reports, but do not fulfill it, because Shopify charges for any shipping labels you buy. On a store that is taking orders, Shopify says you shouldn't turn test mode on, so place one small real order and cancel and refund it, though processor fees may apply. A real order also reaches tools that skip orders Shopify marks as tests.
Does a test purchase change my ad data?
It can. Meta says cleared test activity still appears in Events Manager, Ads Reporting and Audience Creation results, and that server events sent with its test ID are not dropped. Keep test purchases to the few you need.
Does HonestTag send Shopify test orders to ad platforms?
No. HonestTag never sends an order Shopify marks as a test to any ad platform. On paid plans the order's line in Live activity reads not sent: Shopify test order.
Sources
- Shopify Help Center: Placing a test order, read November 6, 2026
- Shopify Help Center: Testing Shopify Payments, read November 6, 2026
- Shopify for Developers: GraphQL Admin API, Order object, read November 6, 2026
- Shopify Help Center: Testing custom pixels, read November 6, 2026
- Meta Business Help: Test events tool: verify app and web browser events for Meta, read November 6, 2026
- Meta Business Help: Differences between event counts in Meta Ads Manager, Ads Reporting and Events Manager, read November 6, 2026
- Meta for Developers: Conversions API, Using the API, read November 6, 2026
- Google Ads Help: About conversion measurement, read November 6, 2026
- Google Ads Help: About conversion windows, read November 6, 2026
- Google Analytics Help: Monitor events in DebugView, read November 6, 2026
- HonestTag docs: server-side delivery
- HonestTag: Order proof
- HonestTag pricing and plan limits
Check your tracking against your real orders
The free Mirror plan shows what Google Ads and Meta claim beside your own orders, and on paid plans every order shows what each platform was sent or why it was not.
- Feed your ads a clean signal. On paid plans, the platforms you send to get each verified purchase, new customers marked as new once confirmed.
- Catch the sales your pixels miss. On paid plans, purchases are sent from your Shopify orders, server to server.
- Know what to scale. From the Truth plan, a weekly scale, hold or cut call when the data supports one, and on every paid plan, proof behind every attributed order.
- Start free. The free Mirror plan puts Google Ads and Meta claims beside your orders. Keep your current tracking while you look.