Someone posted this recently on Reddit:
What's interesting is that most of those answers are actually the same problem in different clothes.
It's not five separate problems. It's one: you're making operational and financial decisions from synced data instead of reconciled data. The dashboards update. The books balance. But the underlying numbers often haven't been verified at the order level.
The gap between what's synced and what's reconciled shows up in five decisions multichannel sellers get wrong all the time, which we’re going to discuss in this guide.
Inventory sitting in a warehouse costs money. Moving it at a discount feels rational. So you pull your margin data, find the products with room, and start discounting.
The question is: what is that margin figure actually measuring?
Gross margin - revenue minus cost of goods - omits everything that happens after the sale is recorded. Platform fees don't appear. FBA fulfillment costs don't appear. Return processing doesn't appear.
A product showing 35% gross margin may be carrying 25–40% in platform and fulfillment costs that sit below that line. After a 20% promotional discount, it doesn't compress margin, it flips negative. You're not moving excess inventory at reduced profit. You're paying to move it.
|
📌Example- A product selling for $40 shows a 40% gross margin, so the team launches a 20% discount campaign. What they don't account for are Amazon referral fees, FBA fulfillment costs, storage fees, and returns. After reconciliation, every discounted sale is actually losing $4. The promotion moves inventory quickly, but over six weeks it quietly drains cash instead of generating it. By the time the actual per-unit economics surface, fees netted, refunds posted, fulfillment allocated, the damage is already compounding. |
If you had order-level reconciliation that nets every fee and refund, the question becomes: which SKUs can absorb 30% vs. which can’t absorb 10%? Discounting decisions become deterministic and measurable.
Amazon: $500,000. Shopify: $300,000. Amazon is growing faster. The revenue line makes the case obvious.
It isn't.
Revenue tells you what was sold. It tells you nothing about what you kept.
Amazon's fee structure includes referral fees, FBA fulfillment, storage, advertising charges, reserve withholdings, and category-specific handling fees, some of which don't appear in Seller Central at the time of sale. They settle later, on a different schedule, against a different period.
When those fees aren't reconciled to the orders that generated them, channel revenue and channel profitability diverge sharply.
|
📌Example- An apparel brand sees Amazon generate $500,000 in revenue versus Shopify's $300,000 and shifts most of its Q4 inventory budget to Amazon. Later, reconciled data reveals Amazon contributed just $20,000 in profit after fees, advertising, and storage costs, while Shopify contributed $45,000. The revenue winner wasn't the profit winner, but the inventory decisions had already been made. |
What that means? Don’t just place your Q4 inventory bets on the first number. The damage won't surface until the settlements come in, weeks after the channel investments are made, the inventory committed, and the ads run.
Your best sellers are moving. Velocity data shows demand. The reorder feels obvious.
Velocity is the right signal, but it’s incomplete. Here’s why!
|
📌Example- A home goods seller reorders a fast-moving SKU selling 200 units per month because velocity reports rank it as a top performer. What the report misses is that the SKU contributes only $2 in profit per unit after referral fees, FBA storage, and other expenses. Meanwhile, a slower-moving product generates $12 per unit and repeatedly goes out of stock. The same working capital in a slower SKU generates more, if you had the data to see it. |
Without fee-adjusted margin at the SKU level, reorder triggers consistently over-stock fast movers and under-stock margin contributors. Dead stock accumulates in 3PL storage.
Working capital gets trapped in units that aren't contributing at the level the velocity number implied. And the SKUs that could actually move the business, the ones with real margin after the full cost stack, go out of stock.
The velocity number was correct. It just wasn't answering the question that determines whether the reorder is a good decision.
A 4x ROAS looks like a win. Platforms report it that way. Agencies lead with it. Growth reviews start with it.
Here's the problem.
ROAS measures revenue per dollar of ad spend. It says nothing about what you kept from that revenue. The result? The campaign starts destroying cash flow.
Platform-reported ROAS doesn't know your fulfillment costs, your return rate, or your per-channel fee stack. It reports what came in, not what you kept. The metric that actually matters is contribution margin per ad dollar, and that number requires order-level data traced through the full cost structure.
Without it, you're scaling campaigns based on revenue generated, not profit generated. Some sellers do this for months before the reconciled numbers surface the gap. By then, the cash is already spent.
|
📌Example- A supplement brand spends $25,000 on ads and generates $100,000 in sales, producing a healthy-looking 4x ROAS. But after fulfillment costs, returns, processing fees, and product costs are reconciled, the campaign is contributing only a few dollars per order. Revenue is growing, but cash flow isn't. The ROAS metric was accurate; it just wasn't measuring profitability. |
A large PO, simple fulfillment, cash coming in. Compared to managing hundreds of individual DTC transactions, wholesale feels clean.
But the decision isn't really about the wholesale order. It's about what that inventory is already committed to.
Most inventory systems show what's on hand, not what's already committed. Future DTC demand, subscriptions, and marketplace sales forecasts often aren't reflected in the inventory available to promise.
As a result, inventory allocated to a wholesale order can leave DTC channels understocked. Products go out of stock, Amazon rankings decline, Shopify conversions drop, and revenue from proven sales channels gets disrupted.
The cash picture compounds the issue. Wholesale orders often operate on Net-30 or Net-60 terms, meaning inventory leaves the warehouse long before payment arrives. What looks like a strong revenue decision can create both inventory pressure and a temporary cash squeeze.
|
📌 Example: |
These aren't five separate mistakes. They're the same mistake five times: a decision made on data that was synced, but not reconciled.
Synced data moves the number from one system to another. Reconciled data verifies that the number is right, every order, fee, refund, payout, and inventory movement accounted for at the transaction level, matched against what actually happened.
The gap isn't obvious in any single decision. A discount campaign that ran six weeks on negative margin.
A Q4 channel bet that grew revenue 38% and profitability 4%. A reorder cycle that trapped $200,000 in inventory that wasn't moving while the margin leaders went out of stock. Each one looks survivable in isolation. The compounding pattern is what breaks businesses.
The data wasn't wrong. It was almost right. And in each of these decisions, almost right was enough to send you confidently in the wrong direction.
Most multichannel operators already have sync. Orders move from Shopify, Amazon, Walmart, or TikTok Shop into QuickBooks automatically. Transactions appear where they're supposed to. The books close.
The problem is that synced data and reconciled data are not the same thing.
Sync moves information. Reconciliation verifies that every order, fee, refund, payout, inventory movement, and settlement actually matches reality. That's where financial visibility comes from.
Almost right is not good enough. Not at this scale, not with this many channels, not with these decisions riding on the numbers.
If you've never looked closely at why the gap exists in the first place, start here.
If you already know something's off and want to understand what fixing it actually looks like, this is the next read.