ERP for Brands Selling on Amazon and Their Own Store

Brands that sell on Amazon and their own store run two operations that behave nothing alike, sharing one pool of inventory. Amazon wants stock positioned in FBA weeks ahead of demand, takes its fees out of a biweekly settlement, and penalizes you for running out. Your own store wants inventory at your 3PL, gives you the customer relationship and the margin, and forgives nothing on shipping cost. An Amazon seller ERP is the system that runs both channels on one inventory, one cost model, and one set of numbers.

Most brands get there the hard way. They start with Shopify plus FBA, add a spreadsheet to plan replenishment, another to reconcile settlements, and a part-time analyst to explain why the two channels' numbers never match. The operational symptoms are predictable: overselling on one channel while overstocked on the other, FBA stockouts that crater organic rank, and a blended margin number that hides which channel actually makes money.

This guide covers why Amazon plus DTC operations break lightweight tooling, what an ERP has to handle to run both channels well, and where general-purpose systems fall short of the workflow.

Running Amazon and DTC together multiplies operational complexity rather than adding it. One inventory pool now serves two demand curves with different lead times: an FBA inbound shipment takes days to weeks from your warehouse to sellable status, while your 3PL picks DTC orders same-day. Allocating stock between them is a standing decision, not a one-time setup, and a spreadsheet updated weekly makes that decision late by definition.

The money side is just as fragmented. Amazon pays out settlements every two weeks with referral fees, FBA fulfillment fees, storage fees, advertising charges, and refunds netted against sales. Shopify revenue arrives through a payment processor with its own fee structure, while shipping costs arrive on carrier invoices. Reconstructing what any single SKU earned across both channels means joining four or five data sources, which is why most brands settle for blended margins and gut feel.

The failure that hurts most is stockout asymmetry. Running out on your own store loses orders; running out on Amazon loses rank, and the sales velocity that took months of advertising spend to build. Brands routinely protect FBA by starving DTC, or vice versa, without a system that shows the cost of either choice.

The requirements are specific to the two-channel workflow, and worth checking against any system you evaluate.

• Unified inventory with channel allocation. One stock record per SKU across your 3PL, FBA, and any 1P warehouse, with rules for how much each channel can sell and visibility into inbound stock at every stage, including receiving at Amazon.
• FBA replenishment as a procurement workflow. Replenishment to FBA behaves like a purchase order against a lead time : it needs reorder points and safety stock math that account for Amazon receiving delays, not just supplier lead times.
• Settlement reconciliation. The system should ingest Amazon settlement reports and tie fees, refunds, and reimbursements back to individual orders and SKUs automatically, and do the equivalent for your store's payment processor.
• Channel-level margin. Real margin by SKU and channel, with Amazon fees, fulfillment costs, and shipping applied per order, so the Amazon vs. DTC investment decision runs on numbers instead of instinct.
• Returns on both paths. FBA customer returns, removal orders, and unsellable inventory on one side; DTC returns through your 3PL on the other, each restocking or writing off correctly.

Demand planning ties these together. Forecasting for Amazon plus DTC operations has to respect that the channels move differently: a Prime Day spike says nothing about your store's Black Friday, and pooling the forecasts produces purchase orders that are wrong for both.

Most ERP systems treat Amazon as just another sales channel, and that assumption fails on contact. A generic connector imports orders, but it does not model FBA inventory as its own location with its own inbound pipeline, does not treat replenishment shipments as first-class transfers with lead times, and does not parse settlement reports beyond posting a lump-sum deposit. The gap gets filled with middleware, spreadsheets, or a hired integrator, and every added layer is another place the numbers disagree.

Legacy mid-market ERPs add a second problem: pace. Amazon changes fee structures, prep requirements, and inbound rules continuously, and marketplace strategy shifts quarter to quarter. When adapting your workflow requires a consultant engagement, the system is perpetually one fee change behind. Point tools built only for Amazon sellers have the opposite gap, handling FBA well while treating your own store, wholesale, and procurement as afterthoughts.

The pattern to avoid is any system that forces a choice between modeling Amazon deeply and modeling the rest of the business deeply. A brand doing meaningful volume on both channels needs both.

Amazon plus DTC is rarely the end state. Brands that succeed on both channels usually add wholesale or retail within a couple of years, and the channel that breaks the system is typically the one you did not plan for. Wholesale brings purchase orders from buyers, case and pallet configurations, EDI documents, and payment terms, none of which resemble either Amazon or DTC order flow. If your operations platform modeled Amazon and Shopify as its entire worldview, the third channel forces either another tool or another migration.

The practical test is how the system models a channel in the first place. A platform that treats channels as configurable order sources feeding shared inventory, costing, and fulfillment logic will absorb a new one as configuration. A platform with hard-coded Amazon and Shopify pipes will not. This is worth asking about while you are still a two-channel business, because the answer determines whether today's system choice is a foundation or the next ceiling.

The same logic applies to selling into new marketplaces or regions. Walmart Marketplace, TikTok Shop, and international Amazon each bring their own fee structures and fulfillment programs, and each is a weekend project or a quarter-long integration depending on the architecture underneath.

DOSS Operations Cloud is a composable operations platform for physical-product brands that runs inventory , orders , and procurement across Amazon, DTC, and wholesale in one system. Native integrations, 70+ of them, connect Amazon, Shopify, 3PLs, and accounting directly, so FBA stock, 3PL stock, and in-transit inventory appear as one governed picture, and settlement data flows back to the orders it belongs to.

The operational payoff is that the two-channel decisions become visible and fast. Operators see sellable stock by location and channel in real time, run FBA replenishment as a workflow with lead times and safety stock built in, and read margin by SKU and channel in DataStudio with fees and fulfillment costs already applied. Because workflows in DOSS are configured, not coded, the response to an Amazon fee change or a new prep requirement is an afternoon's adjustment, not a change order. Verve Coffee Roasters, which sells across DTC, grocery, wholesale, and cafés, cut unbatched orders from 30% to 1% on DOSS and saved 20+ hours a week across its team; Mezcla doubled PO processing speed while saving 12+ hours weekly.

DOSS fits brands roughly $10M-500M in revenue where Amazon and DTC both matter, and where the operations team is done being the integration layer between systems.

Put candidate systems through the two-channel workflow end to end before signing anything. Have the vendor show FBA and 3PL inventory in one view, walk through creating a replenishment shipment with a lead time, and trace one Amazon settlement line back to an order and its margin. Ask what happens when you add a channel or change a workflow: who makes the change, how long it takes, and what it costs. The answers separate systems built for this profile from systems with an Amazon connector on the brochure.

Weight time-to-value honestly. A system that is perfect in month 18 loses to one that is useful in month 2 and complete in month 5, because the spreadsheet regime keeps costing you while you implement.

Selling on Amazon and your own store is a permanent two-front operation: different lead times, different fee structures, different failure modes, one inventory. The brands that run it well stop reconciling between systems and run both channels from one operational truth, with allocation, replenishment, and margin visible in the same place orders live.

DOSS Operations Cloud connects inventory, orders, and procurement in one platform, integrates natively with Amazon, Shopify, your 3PL, and your accounting stack, and gets brands live in months, not years. If your team is still arbitrating between an FBA spreadsheet and a Shopify dashboard, book a demo and see both channels in one system.