SKU Performance Analysis for Shopify: The Complete Guide to Finding Your Winners and Losers
by Trivas.ai
|
9 min read
Oct 20, 2026
SKU performance analysis sounds like something only a data analyst should care about. It's not. If you're running a Shopify store with more than a handful of products, the SKU level is where the actual truth about your business lives. Total revenue can look fine while half your catalog quietly bleeds money. This guide walks through how to actually do sku performance analysis shopify merchants can run without hiring an analyst, where the native reports fall apart, and what to do once you've found the losers.
What SKU Performance Analysis Actually Means
SKU performance analysis means evaluating each individual product variant against revenue, margin, inventory position, and ad spend, not just looking at how a product sold in aggregate.
That distinction matters more than it sounds like. A "product" in your catalog might have eight SKUs under it: four sizes, two colors each. Look at it collection-level and you see one line: 340 units sold, looks healthy. Drop down to the SKU level and you might find that the XL in black barely moves, gets returned constantly, and is quietly dragging down the average.
Here's a concrete version of that. Say you sell a jacket with two SKUs: Medium/Navy and Large/Navy. Same product page, same price, same ad campaign driving traffic to both. Medium/Navy converts well and has a 6% return rate. Large/Navy converts at a similar rate but gets returned 46% of the time, probably a sizing or fit issue specific to that variant. At the product level, this blends into an unremarkable 26% average. At the SKU level, it's an obvious, fixable problem.
This is the foundation everything else sits on. Inventory planning, ad budget allocation, catalog pruning, all of it depends on knowing which specific SKUs are earning their shelf space and which ones aren't.
Why Shopify's Native Reports Aren't Built for This
Shopify's admin will show you sales by variant. That part works fine. What it won't do is connect that sales data to ad spend, contribution margin, or inventory carrying cost in one place.
So merchants end up doing it manually. Export a sales-by-SKU CSV from Shopify. Pull spend data from Meta and Google. Dig up a COGS spreadsheet from wherever it lives, usually an outdated Google Sheet someone built eighteen months ago. Then join all three by SKU, by hand, in Excel.
This works, sort of, at small scale. It falls apart fast past a couple hundred SKUs, or the moment you're selling across more than one channel. VLOOKUP errors creep in. SKU naming gets inconsistent between platforms. Someone forgets to update the COGS sheet after a supplier price increase, and now every margin number downstream is wrong.
This gap, the fact that Shopify tells you what sold but not whether it was worth selling, is exactly why merchants start looking for an analytics layer that sits on top of Shopify instead of next to it. If you're already feeling this pain, it's worth seeing how Shopify data integration can pull order, inventory, and ad data into one place instead of three spreadsheets.
The Metrics That Actually Matter at the SKU Level
Not every metric deserves equal weight here. A few actually move decisions.
Sell-through rate. Units sold divided by units received, over a given period. This is your earliest warning sign for slow movers, well before they turn into dead stock sitting in a warehouse costing you money every month.
Contribution margin per SKU. Revenue minus COGS, shipping, and allocated ad spend. Not gross margin. Gross margin ignores the cost of actually acquiring the sale, which is how a "profitable" SKU turns out to be a net loss once you account for what it cost to get someone to buy it.
Inventory turnover by SKU. How many times a SKU's stock cycles through in a given period. Low turnover ties directly into reorder timing. If you're reordering a SKU that turns over twice a year on the same schedule as one that turns over twelve times, you're mismanaging cash.
Return and refund rate by SKU. This isolates problems that are specific to one variant, size, quality, or a listing photo that doesn't match the product, rather than a problem with the product line overall.
Ad-spend-to-SKU ratio. How much paid traffic is going toward selling each specific SKU. Blended ROAS at the campaign or account level hides this completely. You can have a campaign that looks profitable on average while one SKU inside it is burning spend with nothing to show for it.
Building a SKU Performance Workflow Without a Full-Time Analyst
You need three data sources stitched together: Shopify order and inventory data, ad platform spend data, and a COGS sheet. That's it. The complexity isn't in the sources, it's in keeping the join current.
A reasonable starting cadence: every Monday, pull SKU-level sales and margin for the prior week. Flag anything that falls below a set margin threshold or sell-through rate. Keep the thresholds simple at first, you can tighten them later once you know your catalog's normal range.
Spreadsheets work fine for this under roughly 50 SKUs. One person, one sheet, updated weekly, manageable. Past that, especially once you're running multiple sales channels or warehouses, the manual join becomes the bottleneck. You're spending more time reconciling SKU IDs across platforms than actually analyzing anything.
The real fix is automating the join so it updates daily instead of getting rebuilt by hand every Monday morning. That's less about adding more tools and more about removing the manual step that breaks down as your catalog grows. Teams running this kind of recurring reporting tend to lean on a BI reporting setup specifically so the numbers refresh without someone owning a spreadsheet full-time.
Drilling Down When a SKU Underperforms
Flagging a weak SKU is the easy part. Figuring out why is where most spreadsheet-based processes stall out.
Run through a simple diagnostic sequence: is the problem traffic (not enough people seeing it), conversion (people seeing it but not buying), margin (selling fine but unprofitable), or returns (selling and getting sent back)? Each points to a different fix, and guessing wrong wastes a reorder cycle or an ad budget adjustment.
A real drill-down connects the top-line catalog view straight to the underlying order, ad, and inventory detail for that one variant, without making you re-export anything. Say a SKU shows flat revenue month over month but rising ad spend. On the surface that looks like a demand problem. Drill into the campaign actually driving traffic to that SKU and you might find the real issue: CPC on that specific campaign doubled while conversion rate held steady. The SKU isn't losing appeal, it's just gotten twice as expensive to reach the same number of buyers.
This is the exact workflow the SKU Drill Down module is built around: start at the catalog view, click into the SKU that's underperforming, and get the order, ad, and inventory context immediately instead of hunting for it across three tools.
Common Mistakes That Skew SKU Analysis
A few habits consistently produce bad calls.
Judging SKUs by revenue alone is the most common one. A top-seller by revenue can have near-zero contribution margin once you factor in returns and ad spend. It looks like your best performer and it's actually barely breaking even.
Comparing SKUs across different lifecycle stages is another. A SKU that launched three weeks ago shouldn't be held to the same sell-through threshold as one that's been in the catalog for two years. Early data is noisy by nature.
Ignoring seasonality causes a lot of false alarms too. Flagging a SKU as a loser in its natural off-season, then killing it right before its best quarter, is an expensive mistake that's entirely avoidable with a quick look at last year's numbers.
And margin data goes stale fast. A price change or a supplier cost increase can quietly invalidate your contribution margin numbers. If you're not re-running the analysis after either event, you're making decisions on numbers that were accurate a month ago and aren't anymore.
Turning SKU Data Into Catalog Decisions
The analysis only matters if it changes what you do next. Three decision types come out of it.
Reorder decisions. Combine sell-through rate with supplier lead time. A fast-moving SKU with a long lead time needs to be reordered well before it looks urgent, or you'll stock out. A slow mover with a short lead time can be reordered in smaller batches, more often.
Kill decisions. Set a clear, written rule instead of deciding case by case. Something like: negative contribution margin for two consecutive months means the SKU gets discontinued. A rule removes the sentimental attachment to a product that used to perform.
Scale decisions. Look for SKUs with high margin and low ad saturation, meaning spend could go up without CPCs climbing through the roof. These are your best candidates for a budget increase, and they're easy to miss if you're only looking at top-line ROAS.
None of this should be a once-a-year exercise done during annual planning. Build SKU performance review into a recurring merchandising or planning meeting, monthly at minimum, so the decisions stay current instead of reactive. This is usually a core part of the job for operations managers juggling reorder timing against ad budget calls.
Get SKU-Level Clarity Without the Manual Export Work
SKU performance analysis is only as useful as how often you can refresh it. A perfect spreadsheet model that gets updated once a quarter will always lose to a rougher one that updates daily, because the slow mover you catch in week two is a lot cheaper to fix than the one you catch in month four.
Trivas connects Shopify, your ad platforms, and inventory data so SKU drill down is there when you need it, without the manual joins or the stale COGS sheet. If you're currently stitching this together by hand, it's worth seeing how your Shopify integration data flows into a setup like this before you sink more hours into a spreadsheet process that won't scale past a few hundred SKUs.
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 Saving 10 Hours a Week on Reporting
3 min read
Google Analytics Ecommerce Recommended Events: purchase, add_to_cart, begin_checkout (Official Reference)
3 min read
The Ecommerce Analytics Platform That Replaces Northbeam (Without the Black-Box Attribution)
.d53b12e5.png&w=3840&q=75 "Trivas.ai")
by Trivas.ai
