Ecommerce accounting became its own software category because general ledgers were designed for businesses that sell a thing and collect money for it, and marketplace selling stopped working that way. A single Amazon settlement is not a sale. It is a net wire covering two weeks of gross revenue, refunds, referral fees, fulfillment fees, storage charges, advertising, reserve movements, and sales tax that somebody else already remitted. General accounting software has no opinion about how to take that apart. So a category grew up around the taking apart.
That is the whole story, and it explains why the category keeps expanding rather than getting absorbed back into QuickBooks and Xero.
The settlement is the root cause
Amazon’s own documentation describes the settlement report as a summary of all transactions in a settlement period, including orders, refunds, fees, and adjustments, resolved into a single disbursement. A seller looking at their bank statement sees one deposit. Behind it sit thousands of line items across a dozen categories.
Booking that deposit as revenue is wrong in three directions at once. Revenue is understated because fees were netted out. Expenses are missing because those same fees never got recorded. And the timing is off, because the sales happened across two weeks while the cash landed on one day.
An accountant can unwind this by hand. Many did, for years, and the ones who priced it correctly charged accordingly. What they could not do was make it repeat reliably at volume, across five channels, every two weeks, without errors compounding.
Why the incumbents did not absorb it
The obvious question is why Intuit and Xero did not simply build this. Look at where their inventory functionality sits and the answer becomes clear enough.
QuickBooks Online lists inventory tracking as starting at its Plus plan, priced at $115 per month on Intuit’s US pricing page in August 2026. Simple Start at $38 and Essentials at $75 do not include it. Xero sells Inventory Plus as a paid add-on available only on its Growing and Established plans, per its US pricing page the same month.
Both companies treat ecommerce inventory as an upsell attached to a general accounting product, not as the center of the product. That is a reasonable commercial choice. The overwhelming majority of their customers are service businesses, agencies, contractors, and retailers who do not have units sitting in three fulfillment centers with landed costs that shift by container.
A gap that big invites specialists. The specialists arrived.
The category split in two, and the split is the interesting part
What started as one problem became two products.
The first group solves the settlement itself. A2X describes its job as taking marketplace settlement data and producing accurate summarized journal entries in QuickBooks Online, Xero, or NetSuite, starting at US$29 per month for a single channel according to its pricing page in August 2026. Link My Books occupies similar ground with a stronger tilt toward VAT reporting for sellers dealing with HMRC, per its own pricing page. These tools are narrow on purpose and priced accordingly.
The second group treats the settlement as the entry point rather than the destination. Tools in this group carry inventory valuation, landed cost allocation, SKU-level profit and loss, and reorder forecasting alongside the sync. ConnectBooks sits here, syncing Amazon, Shopify, Walmart, TikTok Shop, and eBay into QuickBooks Online, QuickBooks Desktop Enterprise, and Xero with automated cost of goods sold and SKU-level margin reporting.
The split matters because sellers routinely buy from the wrong group. A $400K seller with forty SKUs buying an inventory platform is paying for machinery they will never start. A $6M seller with eleven hundred SKUs buying a settlement syncer will have correct books and no idea which products are quietly losing money.
Fee complexity is the engine that keeps this running
The category would have stalled if marketplace economics had settled down. They did not.
Amazon’s 2026 US fulfillment fee structure, published in Seller Central, now sets different fulfillment rates across three price bands: under $10, $10 to $50, and above $50. Amazon states that standard-size products priced above $50 see fulfillment fees rise by $0.31 per unit on average versus the prior schedule, while standard-size products priced below $10 rise by $0.05 per unit on average. Amazon also documents a 3.5% fuel and logistics-related surcharge applying to FBA fulfillment fees in the US and Canada starting April 17, 2026, and a holiday peak fee window running October 15, 2026 through January 14, 2027.
Read that again as an accounting problem rather than a pricing problem. The same physical unit now carries a different fulfillment cost depending on what you charged for it, what month it shipped, and whether a surcharge was in effect. A cost of goods sold figure computed from last year’s assumptions is not slightly stale. It is structurally wrong.
No general ledger tracks that on its own. Something has to sit between the marketplace and the books and do the arithmetic every settlement, forever.
Where the category goes next
My read is that the settlement-sync half of the market is close to solved and will compete on price and channel coverage from here. Getting a correct journal entry out of an Amazon settlement is no longer a hard engineering problem, and several products do it well. Margins in that half will compress.
The interesting movement is above the ledger. Sellers do not actually want a correct trial balance. They want to be told that the fulfillment fee band shift pushed three SKUs below break-even in February and nobody noticed until May. That is a different product, and it depends on having SKU-level cost data structured well enough to reason over, which is exactly what the inventory-heavy tools have been accumulating.
The risk is obvious. A tool that tells you what to do is wrong more visibly than a tool that tells you what happened. A misclassified landed cost that produces a bad reorder recommendation costs more than a misclassified landed cost sitting quietly in a report. Sellers should treat confident-sounding output from any of these systems as a hypothesis to check, not an instruction.
What this means if you are choosing today
The category exists because marketplace settlements are genuinely hard to account for, and it persists because marketplace fee structures keep getting more conditional. Neither of those is going away.
Pick based on which half of the category you belong to. If you need correct books and nothing more, the narrow tools are cheaper and will stay cheaper. If your margin question cannot be answered without SKU-level cost data, you are buying an inventory system that happens to sync, and you should price it against the bookkeeper hours it replaces rather than against a $29 competitor doing a different job.