Amazon Ads MCP vs Nova MCP - What Each One Sees 2026
Amazon's Ads MCP server executes campaign changes. Nova MCP reads the whole business, including the fee and COGS context an ads decision needs. Here is the capability-by-capability comparison and the workflow that uses both.
•Amazon's Ads MCP server is an execution layer for the ad account. Nova MCP is a read layer for the whole business.
•Ads data alone cannot answer a margin question: no COGS, no fees, no returns, no inventory.
•Nova MCP covers Sponsored Products down to the customer search term, plus P&L, inventory, listing health and BSR.
•The two are complementary. Read with Nova, act with Amazon.
•Break-even ACoS, not ACoS, is the number that should drive a bid change, and it only exists when fees and COGS are in the model.
Both servers speak the same protocol and connect to the same assistant, which makes them look interchangeable. They are not. One changes your campaigns, the other tells you whether the change makes money. Here is exactly what each one sees.
A PPC spend analysis shows the kind of question a connected AI assistant can investigate.
What Amazon's Ads MCP server sees
Amazon's Ads MCP server exposes the advertising account through the Model Context Protocol. An assistant can pull campaign performance, create campaigns, and adjust targeting and bids without anyone writing API code. If your question is "set up a new exact-match campaign for these keywords", this is the right tool and it is genuinely good at it.
Its boundary is the ad account. It reports ad spend and ad-attributed sales. It does not know what a unit costs you, what Amazon charged you in referral and fulfilment fees, how many units came back as returns, or how much stock is left. Those numbers live in Seller Central and in your own cost setup, not in the ads API.
What Nova MCP sees
Nova MCP exposes a modeled seller dataset across 23 marketplaces with real-time sales and ads refreshed every two hours. Advertising is one surface among several: Sponsored Products at campaign, ad group, targeting, keyword and attributed SKU grain, plus the actual customer search queries behind the spend. Around it sit the P&L (40+ Amazon fee types reconciled at SKU level), FBA and AWD inventory with days of cover, current listing health and buy box state, and Best Seller Rank.
The practical difference is what the assistant can conclude. Ads data says a keyword has a 45% ACoS. The joined model says that keyword sits above break-even on a product whose contribution margin is 28%, so the spend is destroying profit even though the campaign looks busy.
Amazon Ads MCP vs Nova MCP, by capability
Capability
Amazon Ads MCP
Nova MCP
Campaign and bid changes
Ad spend and ad-attributed sales
Customer search-term detail
Referral, FBA and storage fees
COGS and contribution margin
Returns and reimbursements in the P&L
FBA and AWD inventory cover
Listing health and buy box state
Best Seller Rank
Organic sales, so TACoS is a real number
Marketplaces
Per ad account
23 marketplaces in one model
Access model
Read and write
Reads by default; confirmed listing edits only
ACoS is not a profit metric
An ads-only assistant optimises the number it can see. That number is a ratio of ad spend to ad revenue, before every cost Amazon charges you. Break-even ACoS differs per SKU because fees and COGS differ per SKU. Without the fee model in context, "reduce ACoS" is advice with no connection to your bank balance.
Give your assistant the profit context ads data is missing
Nova MCP joins spend, fees, COGS and inventory so a bid decision has a margin answer behind it.
Enable both servers in the same assistant and the loop closes:
Diagnose with Nova. "Which search terms spent more than $100 last month with zero sales?" comes back as a ranked table in one call, filtered server side.
Check the profit side, still with Nova. "For the SKUs behind that spend, what is contribution margin after fees?" tells you whether to cut the term or fix the price.
Act with Amazon. Ask the Ads MCP server to add the negative keywords or move the budget.
Nothing in that loop requires a spreadsheet, and nothing in it asks the model to invent a fee calculation. Our PPC analytics feature page covers the same data in the cockpit, the Seller Central MCP page compares Nova with Amazon's new Selling Partner plugin, and the Nova MCP tools page lists every tool the assistant can call.
Why the split exists at all
Amazon builds against its own surfaces. The ads API is a clean, well-scoped product, so an ads MCP server is a natural extension of it. A profit MCP would require Amazon to model your cost of goods, your inbound freight, your VAT treatment and your multi-account structure, which are things Amazon does not hold.
That is also why community wrappers around the Selling Partner API do not close the gap. They expose endpoints; they do not reconcile them. Amazon documents strict per-operation rate limits in its SP-API usage plans reference, so an assistant paging through raw settlement data hits throttling long before it produces a margin figure.
Our take
Which one do you actually need?
If you only execute ad changes, Amazon's server is enough. If you decide budgets, prices or restocks, you need the profit layer underneath it.
Best fit if
•Sellers whose ad decisions depend on margin, not just ACoS
•Agencies that report profit to clients, not just ad performance
•Teams running both a read and an action server in one assistant
Skip if
•Pure campaign operators with no P&L responsibility
•Teams that need campaign, bid, price or stock changes from the assistant
It is Amazon's own MCP server for the advertising account. It lets an AI assistant work with Sponsored Ads campaigns: pull reporting, create campaigns, adjust targeting and bids, through a standard protocol interface rather than direct API code.
Anything outside the ad account. It has no view of COGS, referral and FBA fees, returns, reimbursements, storage costs, inventory levels or organic sales. That means it cannot tell you whether an ad-efficient SKU is actually profitable.
No. They solve different halves of the job. Amazon's server executes changes on the ad account. Nova MCP reads the whole business, including the profit context ads decisions need. Most teams enable both in the same assistant.
Yes. MCP clients allow several servers at once and the assistant picks the right tool per question. A natural flow is to ask Nova which keywords waste money, then ask the Amazon Ads server to apply the negative keywords.
No. Nova MCP does not change campaigns, budgets or bids. It reads by default, and optional writes are limited to listing edits that a workspace owner enables, previews and confirms. Prices and stock levels are never changed.
Because ACoS is a revenue ratio, not a profit ratio. Two SKUs at the same ACoS can sit on opposite sides of break-even once referral fees, FBA fees, returns and COGS are counted. Without the fee model, an ads-only assistant optimises toward the wrong target.
Ready to Transform Your Amazon Business?
Join thousands of successful sellers who use Nova Analytics to make data-driven decisions and maximize their profits.
