What is MCP?
MCP stands for Model Context Protocol — an open standard that lets AI models like Claude and ChatGPT connect directly to external tools and data sources. If an API connects one piece of software to another, an MCP connects your software to an AI model. Once connected, that tool becomes a data source the model can query in real time, in response to plain-language questions.
The Teikametrics MCP applies this to your marketplace advertising and catalog data, built on our platform's history of processing over $10 billion in marketplace transactions.
What you can use it for
Once connected, you can ask your model plain-language questions instead of building dashboards or exports. A few examples:
"Calculate blended contribution margin across my full catalog. Flag any ASIN where ACOS looks healthy but margin is negative."
"Compare my top 10 products across Amazon, Walmart, and TikTok Shop. Flag any price-parity risk, stockouts, or aging inventory."
"Sponsored Products revenue dropped 18% this week on this ASIN — check catalog changes, inventory, competitor rank, and bid changes, and tell me the most likely cause."
"Audit my listings against our brand guidelines and Amazon's latest content policy. Flag anything out of compliance."
Because the model reasons across every connected data source in one conversation, questions that used to require exporting from three systems and doing VLOOKUPs in Excel can now be asked directly, live.
How a query actually works
You ask a question in plain language — for example, "Compare my top 10 products across Amazon, Walmart, and TikTok Shop, and flag any price-parity risk."
The model identifies which of the 12 tools it needs to answer the question.
It queries that data live, scoped to your login's permissions — not from a cached export.
It reasons across the results — merging, comparing, and calculating the way an analyst would.
You get a structured answer, with the option to ask for the underlying formula and inputs so you can verify the work before acting on it.
What data is available?
The Teikametrics connector gives read-only access to advertising performance, SKU-level catalog profitability, and listing-content data for the advertisers your login can reach. It exposes 12 tools in four groups.
Data area | Tools | What you get |
Accounts | 1 | The list of advertiser profiles you can query |
Advertising performance | 8 | Campaign, ad group, product ad and target metrics, each as a detail view and a rolled-up aggregate |
Product catalog | 1 | One row per SKU, blending ad, organic, profit, COGS and inventory data |
Listing content | 2 | Smart Pages optimization state, plus live and AI-generated listing text |
Everything here is read-only. The connector cannot change bids, budgets, COGS or listings.
Accounts and advertisers
The account list returns several hundred advertiser profiles on Amazon and Walmart; no TikTok advertisers appear even though some tools accept TikTok ad types. Every other tool needs an advertising_id from this list.
Field | What it tells you |
name | Profile name (a few are blank) |
advertising_id | The ID every other tool requires |
channel | amazon or walmart |
account_name | The owning account, such as Fruit of the Loom or Moose Toys |
country_code | Marketplace country, such as US, CA, MX, GB, DE or JP |
Things to know when picking an advertiser:
Amazon profiles are often split by country, so one brand can have a separate ID for the US, Canada, Mexico and each European marketplace.
Walmart profiles are US only, and many brands have separate 1P and 3P profiles.
Some profiles are demo or test entries (Sales Demo Account, Shout About), or are labelled "DO NOT USE" or "Not Advertising".
A few advertising IDs appear under more than one account, so confirm the account name before querying.
Advertising performance
Advertising data comes at four levels, from campaign down to individual targets. Each level has a detail tool that returns rows with pagination and an aggregate tool that returns one rolled-up summary.
Level | Detail tool returns | Aggregate tool |
Campaign | Name, ID, status, impressions, clicks, spend, sales, ACOS, ROAS, CTR, CPC, and the AI-bidding goal | Totals across all campaigns |
Ad group | Name, IDs, status, impressions, clicks, spend, sales, orders, ACOS, ROAS, plus bidding goal, ACOS limit, and min and max bid | Totals across ad groups, optionally for one campaign |
Product ad | Product title, ASIN or item ID, ad ID, ad state, impressions, clicks, spend, sales, orders, ACOS, ROAS | Totals across product ads, optionally for one campaign or ad group |
Target | Target text or expression, match type, bid, status, ad group, campaign, plus impressions, clicks, spend, sales, ACOS, ROAS, CTR, CPC, units, conversions, conversion rate | Totals for keyword or product targets, optionally for one campaign or ad group |
Aggregate metrics: impressions, clicks, spend, sales, ACOS, ROAS, CTR, CPC, units sold, ad conversions and conversion rate, with the covered start and end dates.
Ad types: sponsoredProducts and sponsoredBrands (Amazon and Walmart), sponsoredVideo (Amazon and Walmart), sponsoredDisplay (Amazon only), and the TikTok types productSales, GMVMaxProduct and GMVMaxLive.
Bidding settings: the seller's AI-bidding goal is returned at campaign and ad group level. The ACOS limit, expressed as a fraction (0.25 means 25%), and the min and max bid are set per ad group and may be empty. Goals include grow, profit, roas, launch, liquidate, scheduled and external; "external" means AI bidding is off.
Filters: campaigns can be filtered by goal, and every lower level can be narrowed by campaign name or ad group name. Product targets are Amazon only.
Product catalog performance
The catalog tool returns one row per SKU and blends advertising, organic sales, profitability and inventory. It works for Amazon, Walmart and TikTok, and can be narrowed to a single ASIN or item ID.
Group | Fields |
Identifiers | SKU, SKU name, variation type, parent SKU, external item ID and type, base item ID, sales channel, currency, advertised status |
Sales | Total sales, total units sold, average order price |
Profit | Estimated gross profit, estimated gross margin, pre-ad gross margin, estimated fees |
COGS | Unit COGS (as configured by the seller), aggregate COGS for units sold in the window |
Advertising | Ad sales, ad spend, ad conversions, ACOS, TACoS, ROAS, ad conversion rate, impressions, clicks, CTR, CPC, campaign count |
Inventory | Current inventory quantity, inventory value |
Metrics are aggregated over the chosen window, not returned as a daily series. Margin fields come back empty when no unit COGS has been entered for the SKU.
Listing content
Two tools cover listing content: one shows where each product stands in Smart Pages, and the other shows the actual text.
Listing optimization data returns one row per enrolled product:
External product ID and product name
Enrollment status (SMART or UNENROLLED)
Optimization state, the raw Smart Pages status
Recommendation types ready for the product (today only text_keyword)
Whether a recommendation is ready to generate
Last optimization date and state change date
It can be filtered by optimization state, for example OPTIMIZE_COMPLETED for content ready to review, or by listings that changed within the last 7, 14, 30 or 90 days.
Generated content takes up to 50 product IDs per call and returns, for each:
The current live listing text
The AI-generated recommendation
Any user edits, which are often empty
Targeted keywords, split by placement
IDs with no matching listing, or whose fetch failed, are reported separately and do not block the rest.
Query inputs and data caveats
Most tools share the same inputs, and a few data quirks affect how results should be read.
Input | Values |
advertiser_id | The advertising_id from the account list |
marketplace | amazon, walmart or tiktok (catalog tool) |
lookback_window | 7d, 14d, 30d or 90d |
advertising_type | Required for all advertising tools (see ad types above) |
cursor | Pagination token; each detail result returns a next cursor until the last page |
Reporting lag: the window ends 1 day before today for Amazon and 2 days before today for Walmart.
Currency: monetary amounts are reported in USD.
Aggregation: metrics cover the whole window; there is no day-by-day breakdown.
Margins and COGS: both depend on COGS being entered in Teikametrics, and margins are estimates based on its fee calculations.
Ad group names can repeat across campaigns, so pass the campaign name as well when filtering.
What is not available
The connector cannot write anything, and it does not return several data types that are often needed for root-cause work.
Changes: no edits to bids, budgets, goals, COGS or listings.
Time series: no daily or weekly trend data; the longest window is 90 days.
Search terms: no search-term report, only keyword and product targets.
Operational context: no fulfillment method, Buy Box, organic rank, price history or inventory history; inventory is a current snapshot.
Listing assets: no images, A+ content, ratings, reviews or competitor listings.
Other channels: advertiser profiles are Amazon and Walmart only.