Why Northbeam Shows Different Numbers Than Shopify (And How to Reconcile Them)
by Trivas.ai
|
7 min read
Sep 08, 2026
Pull up your Shopify orders for yesterday. Now pull up Northbeam. Unless you're running zero paid spend, those two numbers don't match, and they're probably not close. This trips up almost every DTC team that adopts Northbeam, and the reason isn't a bug. It's baked into how the two systems count revenue in the first place. If you've been searching for why Northbeam shows different numbers than Shopify, the short answer is that they're not measuring the same thing, even though the dashboards look like they should be.
Why Northbeam and Shopify Almost Never Match Exactly
Shopify counts every order placed on your storefront. Doesn't matter if it came from a paid ad, an email, a Google search, or someone typing your URL from memory. If the order exists, it's in the total.
Northbeam only counts what it can attribute. That means an order needs a tracked click or a matched view sitting inside its lookback window. No click ID, no pixel match, no attribution. The order still happened, Shopify still has it, but Northbeam has no way to tie it to a channel.
Here's what that looks like in practice. Say Shopify shows 850 orders for the day. Northbeam attributes 610 of those to paid channels. The other 240 land in organic, direct, or straight up unattributed. That's not Northbeam malfunctioning, that's Northbeam doing exactly what it was built to do: track what it can see, and leave the rest alone.
This delta has a name in the industry: the attribution gap. It's the difference between total store revenue and the sum of everything your ad platforms and attribution tools claim credit for. Every brand running paid social has one. The question isn't whether it exists, it's whether the size of it makes sense for your setup. Anyone managing Shopify analytics alongside a third-party attribution tool needs to expect this gap as a baseline, not a warning sign.
Attribution Model Differences That Create the Gap
Even inside Northbeam, the model you pick changes the math. Last-click attribution hands the whole order to the final touchpoint. Multi-touch spreads credit across the path. Data-driven models weight touchpoints based on modeled influence. Same order, three different revenue splits, and none of them will match a Shopify total that doesn't care about touchpoints at all.
Lookback windows add another layer. Northbeam's default is 30-day click, 1-day view. Shopify just timestamps the order the moment it's created. So a customer who clicked an ad 25 days ago and bought today shows up in Northbeam's window but arrived in Shopify with zero context about that click.
View-through attribution is the one that trips people up most. It lets Northbeam count a conversion from someone who saw an ad impression and never clicked anything. That's a real methodology choice, but it means attributed revenue can include people who might have converted anyway.
Then there's cross-device behavior. Someone browses on their phone during lunch, buys from their laptop that night. Unless Northbeam has a matched identity linking those two sessions, it sees two separate, incomplete journeys. Shopify sees one order and doesn't care how the customer got there.
Order-Level Timing and Financial Definitions
Shopify counts an order the second it's placed. Pending payment, unpaid, even later cancelled, all of it counts unless you've gone in and applied filters. That inflates the Shopify side relative to what actually shipped.
Northbeam generally reports revenue net of refunds and cancellations. It may also exclude subscription renewals or POS orders entirely, depending on how it's configured. So you're comparing a gross, everything-included number against a cleaner, filtered one, and wondering why they don't line up.
Timezones make this worse than most people expect. Shopify defaults to your store's local timezone. Northbeam dashboards can default to UTC or the account timezone instead. An order placed at 11:40pm local time can land on a completely different calendar day depending on which system you're reading. Run a day-over-day comparison without checking this, and you'll chase a discrepancy that's really just a clock mismatch.
Tracking Loss on the Ad Platform Side
iOS 14.5+ and Safari's Intelligent Tracking Prevention block a real chunk of the pixel fires Northbeam depends on to match clicks and views to conversions. This isn't a Northbeam-specific problem, every attribution tool built on browser tracking deals with it, but it's a direct contributor to the gap.
Ad blockers and cookie consent banners cut into trackable sessions before Northbeam ever gets a chance to see them. If a user declines cookies or blocks a script outright, that session is invisible from the jump.
Server-side tracking helps close some of this. It routes conversion data through your server instead of relying purely on browser-side pixels. But it doesn't get you to 100% match rates, not for anyone. Real orders, placed by real customers who clicked real ads, can end up permanently unattributed simply because the tracking chain broke somewhere along the way.
How to Reconcile the Two Numbers: A Checklist
Before assuming something's broken, run through this:
Match the date range and timezone. Pull both platforms using your store's actual local timezone, not whatever UTC default is sitting there. This alone resolves a surprising number of "why don't these match" tickets.
Compare gross vs net definitions. Check whether refunds, cancellations, taxes, and shipping are included on each side. If Shopify includes gross sales and Northbeam reports net, you're not looking at the same base number to begin with.
Check the attribution window against your actual buying cycle. A 30-day default window makes sense for an impulse purchase, not for a $400 mattress people research for three weeks. If your buying cycle is longer, extend the window or expect a bigger gap.
Treat unattributed orders as expected, not broken. Orders with no UTM parameter and no click ID aren't proof of a tracking failure. They're often word of mouth, branded search, or someone who just remembered your name. Call this "dark" revenue and stop trying to force an explanation onto every dollar of it.
Teams setting this up for the first time should also check their Shopify integration settings directly, since a lot of reconciliation issues trace back to how order data syncs in the first place.
When the Gap Is a Red Flag vs Normal
A 15-30% delta between platform-attributed revenue and total Shopify revenue is normal for a DTC brand spending meaningfully on paid social. Don't panic at that range on its own.
What should worry you: the gap widening month over month with no change in your spend mix. Same channels, same budget allocation, growing gap. That pattern usually means something broke in the tracking implementation, not that customer behavior shifted overnight.
The other flag is the opposite problem: attributed revenue across your ad platforms summing to more than total Shopify revenue. That's double-counting, and it's common with multi-touch models where more than one channel claims full credit for the same order. If your Meta, Google, and TikTok numbers added together exceed what Shopify actually processed, that's not a coincidence worth ignoring.
Reconciling at the Data Layer Instead of the Dashboard Layer
Most of this mess exists because attribution tools and store platforms live in separate systems, each reporting its own version of the truth with no shared source underneath.
Trivas builds its reporting on top of Amazon Redshift, pulling raw Shopify order data alongside ad platform spend and conversion data into one warehouse. That keeps the revenue baseline tied to what actually happened in your store, instead of whatever a dashboard's default filters decide to show you. It's part of the BI reporting layer, and it's the reason reconciliation stops being a manual monthly chore.
The Wingman AI layer sits on top of that and flags when attributed revenue and Shopify revenue diverge past a threshold you set, instead of leaving a founder to notice the gap three weeks late while scrolling through two open tabs.
If you're actively comparing Northbeam against other options, the breakdown at Northbeam vs Polar vs Trivas covers where each tool draws its attribution lines differently.
The Takeaway
A gap between Northbeam and Shopify is, almost always, a definitional mismatch. Different lookback windows, different gross/net rules, different timezones, different tracking loss. None of that means your spend is wasted or your pixel is broken.
Run the reconciliation checklist before you assume the worst. Match your date ranges, check your revenue definitions, confirm your attribution window actually fits your buying cycle. Most "why Northbeam shows different numbers than Shopify" panics resolve the moment someone actually checks these settings side by side.
If you'd rather see order and attribution data reconciled in one place instead of doing this by hand every month, take a look at how Trivas puts it together, or just keep an eye on the blog for more of this kind of breakdown.
Content author and contributor at Trivas.ai, sharing insights on e-commerce analytics, business intelligence, and data-driven strategies to help businesses grow.
Continue Reading
explore more insights
Ecommerce Analytics for Pet Brands on Shopify and Amazon
3 min read
Trivas.ai Free Trial: What You'll Actually See in Your First 7 Days
3 min read
How to Forecast Q4 Ecommerce Revenue Using AI (Without Guessing)
.d53b12e5.png&w=3840&q=75 "Trivas.ai")
by Trivas.ai
