by Trivas.aiMost Shopify brands can tell you last month's revenue down to the dollar. Ask them what share came from new customers versus repeat buyers, and you get a shrug. That gap matters more than it looks. Shopify analytics with new vs returning customer split is the difference between knowing you grew and knowing why you grew, or whether you're actually growing at all.
Revenue and AOV get the spotlight because Shopify surfaces them front and center. The split between new and returning customers takes more digging, so most teams skip it.
That's a problem. Blended metrics smooth over cracks. A brand can hit its revenue target for the quarter while new customer acquisition quietly drops 15%, propped up entirely by returning buyers spending more per order. The topline chart looks fine. The engine underneath is stalling.
This isn't a nice-to-have for data analysts. Founders and marketing leads need this split before they decide where the next dollar of budget goes. Pour more into paid acquisition when the real issue is retention, and you'll pay for clicks that don't fix the problem. Cut ad spend when new customer flow is actually healthy, and you choke growth that was working fine.
Shopify does have a "New vs Returning Customer Sales" report tucked into the analytics section. It shows order count and sales value split by customer type. That's it.
No connection to ad spend. No channel breakdown. No way to see if new customer revenue came from Meta, Google, TikTok, or organic search. You get a number, not context.
Want to know your CAC for new customers specifically, not blended CAC across your whole customer base? Not in this report. Want to see repeat purchase behavior by product line? Also not here.
So merchants do what merchants always do: export the CSV, open Excel or Google Sheets, and build a pivot table by hand. Every week, or every month, someone recreates the same manual join between Shopify data and ad platform spend. It works, until someone's on vacation or the spreadsheet breaks.
A split worth acting on needs more than a two-column table. Here's what should actually be in it:
Revenue and order count by customer type, filterable by date range, product, and marketing channel. Not a static report, a live view you can slice.
New customer CAC, calculated from actual ad spend across Meta, Google, and TikTok, divided by actual new customers acquired, not total orders. Blending the two overstates how efficient your acquisition really is.
Repeat purchase rate and average time between orders for returning customers. This separates people who bought twice by coincidence from people who are genuinely loyal.
A cohort view, showing what a given month's new customers do in the following 30, 60, and 90 days. Are they coming back? How fast? That's the leading indicator most brands never look at until it's too late.
Trivas pulls Shopify order and customer data into Redshift and joins it with ad platform spend at the customer level, not just the order level. That distinction matters: order-level joins tell you what sold, customer-level joins tell you who bought and whether they'd bought before.
From there, dashboards separate new customer CAC from blended CAC automatically. You see the real cost of acquiring a first-time buyer, broken out by channel, instead of one averaged number that hides which platform is actually working.
Wingman, the AI insights layer, watches for shifts like "new customer share dropped 12% week over week" and surfaces it without anyone building that report by hand. That's the part most teams are missing today, not the data itself, but someone catching the trend before it shows up in a board deck three months later.
Setup runs off your existing Shopify integration. No manual customer list uploads, no separate tagging system to maintain. If you've already got Shopify connected through the Shopify integration, the split is mostly there waiting for you.
Here's where the split earns its keep. Two scenarios, same-looking revenue chart, completely different problems.
Scenario 1: Revenue is flat. New customer share is rising. Repeat purchase rate is falling. This is a retention problem wearing an acquisition costume. You're bringing in enough new buyers, but they're not sticking. The fix is CRM, email flows, loyalty programs, not more ad spend.
Scenario 2: Revenue is up, but almost entirely from returning customers. Total sales look healthy, so nobody's panicking. But paid acquisition is quietly underperforming, and the growth you're celebrating is really just your existing base spending more. Turn off the ads tomorrow and the top-line number keeps looking fine for a while, then it doesn't.
Each of these should change the very next budget conversation. Scenario 1 means shifting dollars toward retention. Scenario 2 means auditing your acquisition channels before you assume they're doing their job. Neither is visible without the split. Both are exactly why marketing leaders need this view before, not after, the quarterly planning meeting.
Getting this running isn't a multi-week project.
Start by installing Trivas AI on the Shopify App Store to connect your order and customer data. Then connect your ad platforms, Meta, Google, TikTok, so CAC and ROAS split by customer type automatically instead of getting blended into one number.
Most stores are up and running in under a day. It reuses connections you've already got: Shopify, your ad accounts. No custom event tracking, no pixel work, no waiting on a dev to fire new tags.
If you're weighing this against a broader BI setup, BI reporting covers how the dashboards handle this alongside your other channels, not just Shopify in isolation.
Blended metrics hide the one thing that actually tells you whether your business is growing or just treading water: is this acquisition-driven growth, or retention-driven growth? Shopify's native reports won't connect that split to spend, so most brands never look past the surface number.
If you want to see your own store's new vs returning split against real ad spend, start a trial and it'll pull in within minutes, not a multi-week onboarding.
This is also just the starting point. Once you've got the split, the next layer is cohort analysis and LTV by channel, both worth digging into once you've got the basics running.
explore more insights
3 min read
3 min read
3 min read
.d53b12e5.png&w=3840&q=75 "Trivas.ai")