by Om RathodPicture a normal week for a mid-size DTC brand: tiered volume discounts running on the site, subscriber pricing knocked off for repeat customers, a BOGO promo live for a product launch, three influencer codes stacking on top of each other, and a seasonal bundle deal layered over all of it. None of it reconciles in one dashboard. Ad platform says ROAS is strong. Finance says margin is shrinking. Nobody can say why with confidence.
That's the symptom worth naming directly: blended ROAS looks great in Meta or Google's dashboard, but net margin tells a completely different story once every discount type gets applied. The ad platform sees revenue. It doesn't see that half of that revenue came in at 35% off.
This is written for DTC brands on Shopify and/or Amazon doing enough volume that promo complexity has stopped being a pricing decision and started being a reporting problem. Running two discount types, you can probably still track it by hand. Running five, you need ecommerce analytics for a brand with complex discount structures: a data layer that tracks every discount type back to SKU-level margin, not just top-line revenue.
Most ecommerce analytics tools on the market, Triple Whale, Northbeam, Polar, were built around a simpler assumption: one discount code, one order, one clean revenue number. That works fine for a brand running the occasional 10% off email code. It falls apart the moment you're stacking promo types.
Here's the specific failure mode. Tools that pull gross revenue straight from the Shopify Orders API often can't separate automatic discounts, code-based discounts, and Shopify Scripts or Functions-based tiered pricing into distinct line items. They see a discounted total. They don't see which discount logic produced it, or how much of that discount was automatic versus code-triggered versus tier-based.
That creates a real margin blindspot. A $40 AOV order with a 30% bundle discount stacked on a 15% subscriber discount shows up looking identical to a full-price order, unless the platform is specifically tracking discount type and depth as separate fields. Revenue looks the same. Margin isn't even close.
The manual workaround most teams land on: exporting orders to spreadsheets every week and manually backing out true margin by promo type. It works, technically. It also adds hours to a reporting cycle that should take minutes, and it breaks the first time someone launches a new promo type nobody built a spreadsheet formula for yet.
Solving this isn't about a nicer dashboard. It's a different data model underneath the dashboard. A handful of things have to be true:
Line-item level discount attribution. Every order needs to break down by discount type, tiered, BOGO, bundle, code-based, automatic, not collapse into a single generic "discount total" field. If you can't see which discount type touched an order, you can't isolate its margin impact.
SKU and variant-level margin after discount. Order-level margin hides too much. A bundle might carry three SKUs with three different cost structures. You need margin visibility down to the variant, post-discount, not a blended order average.
Cross-channel reconciliation. A bundle discount running on Shopify and a lightning deal running on Amazon need to land in the same margin view. Otherwise you're comparing channel performance on apples-to-oranges promo math.
Time-based promo performance. You need to isolate a specific campaign window, say a 72-hour tiered flash sale, and see true incremental margin for that window specifically, not just a revenue lift number that ignores what the discount cost you to generate it.
This is the actual bar for ecommerce analytics for a brand with complex discount structures. Anything short of these four just moves the manual spreadsheet problem one step downstream.
Trivas is built around the idea that discount data shouldn't get flattened before it reaches your dashboard. Raw order and discount data gets ingested into Amazon Redshift with discount type, code, and stacking order preserved, not collapsed into one number the moment it hits the database.
From there, BI reporting dashboards let teams filter P&L views by discount category. A tiered volume promo and a subscriber discount show up as two separate cost lines instead of one blended figure. That's the difference between "discounts cost us $12,000 this week" and "the tiered promo cost $8,200, the subscriber discount cost $3,800, and here's which one is actually worth keeping."
The Wingman AI insights layer sits on top of that and does the flagging work automatically. Instead of digging through filtered views yourself, it surfaces things like BOGO orders running 40% lower contribution margin than the campaign's blended average, before that gap quietly eats a quarter's profit.
And because the underlying data already separates discount type from revenue, forecasting and simulation can model a new tiered discount structure against historical margin data before it ever launches. Test whether a proposed three-tier volume discount actually pencils out, instead of finding out three days after the sale ends that it didn't.
Triple Whale, Northbeam, and Polar are all genuinely strong at ad attribution and MER. That's not in question. What's less clear is whether any of them were built with discount-type-level margin breakdowns as a core part of their data model, versus bolted on as a secondary reporting field.
[VERIFY]: specific claims about how each competitor handles stacked or tiered discounts should be confirmed directly against their current documentation before publishing. We're not asserting a feature gap here that hasn't been checked, just flagging that this is the area worth digging into if you're evaluating them for a promo-heavy brand.
If you're actively weighing Trivas against Northbeam and Polar, the Northbeam vs. Polar vs. Trivas comparison breaks down the differences in data depth and customization in more detail.
Here's the honest positioning: Trivas is the right fit when your core problem is margin visibility across complex promo structures, not simply attributing ad spend to revenue. If your discount stack is simple, any of these tools will probably serve you fine. If it's not, the data model underneath matters more than the dashboard on top of it.
Run through this quickly:
If two or more of these are yes, standard ecommerce analytics tools are already costing you visibility. Not eventually. Right now, every week you close out reporting without knowing which discount actually ate your margin.
If tiered promos, BOGO, bundles, and code stacking have made your margin reporting a weekly spreadsheet exercise, that's exactly the problem Trivas was built to solve. Stop guessing at blended margin. See exactly what each discount type costs you, broken down by SKU, channel, and campaign window.
Onboarding starts by mapping your existing discount rules, Shopify Functions, Amazon promotions, code-based discounts, directly into the Trivas data model, so the breakdown is accurate from day one rather than something you have to configure after go-live.
Start a trial or talk to the founding team about your specific discount structure before your next promo goes live.
explore more insights
3 min read
3 min read
3 min read
.d53b12e5.png&w=3840&q=75 "Trivas.ai")