On 10 November 2026 Anthropic put three new work surfaces into Claude at once. Claude Dashboards, in beta since October, left beta under a new name, Claude Data. Claude Sheets and Claude Whiteboards arrived next to it, both in beta. All three are on every paid Claude plan, on the web, on desktop and on mobile.
For someone who sells on Amazon, the news is less about the three names and more about one sentence in the announcement: Claude Data needs a data connector. A dashboard is only as wide as the data behind it. This post goes through what each product does, which of the three is worth an Amazon team's time this week, the two routes Amazon data takes into Claude Data, and what to set up first.
Claude Data is the new name for Claude Dashboards. You ask a question about your company's numbers in plain language, Claude writes a query for each chart, runs it against your data source and lays the result out as a dashboard. Every chart shows the query behind it and when its data was last refreshed. The renamed product adds several things on top of that:
Anthropic names Amazon Redshift, BigQuery and ClickHouse among the warehouses it reads, and says a CRM can be a source too.
Claude Sheets builds and edits spreadsheets with tabs, formulas, formatting and drop-downs, often from one prompt. Each edit Claude makes is outlined so you can follow what changed, and a sheet can be shared so a team and Claude work in it together. Anthropic is plain that the functionality is simple for now. For the full Excel feature set it points to exporting as .xlsx or CSV, or to Claude for Excel.
Claude Whiteboards is a shared board. You talk a plan through and Claude turns it into a timeline with owners and dates. Teammates add notes at the same moment, Claude groups them by theme and drafts action items, and you can drag a box around any part of the board and ask about it. A finished board can become a doc or a deck, or move into Lucid, Miro or Vani by Zoho.
The same day, Anthropic also widened interactive connectors: tools such as Notion or Confluence can now open as full apps beside a conversation. That is a story for another post.
Claude Data matters most. Amazon sellers and vendors have asked for one thing from every AI tool since the first connector appeared: a dashboard they did not have to build. Sales by marketplace, ad spend against profit, stock cover, purchase orders. Claude Data is the first time Claude produces that as a living page with the query visible under each number, instead of a chart that is gone when the chat ends.
Claude Sheets is useful, in a narrower way. A lot of Amazon work still lives in trackers: a reorder plan, a list of search terms to negate, a launch checklist with dates. A sheet that Claude fills and that the team then edits by hand suits that work better than a dashboard does. Keep in mind that Sheets is the simple version today.
Claude Whiteboards matters least, and that is fine. Its one clear use for an Amazon team is the weekly business review or a launch plan: notes from four people, grouped, with owners and dates. It is a meeting tool, and it does not need Amazon data to earn its place.
Anthropic's wording is that Claude Data requires a data connector, such as a warehouse or a CRM, and that it works with the apps you have connected to Claude. Amazon does not appear in the list of named sources, so the data has to arrive through something. With DataDoe there are two routes.
In Claude, DataDoe is a custom connector: you open Customize → Connectors → Add and paste the server address. In an ordinary chat, Claude can then query Seller Central, Vendor Central and Amazon Ads as one schema, including past periods and the product costs you upload. The setup is on our page about connecting Amazon to Claude.
In a dashboard’s Sources menu, the live sharing switch sits next to each MCP connector, and an MCP connector is exactly what DataDoe is in Claude.
The second route is documented on both sides. Anthropic names BigQuery as a Claude Data source. DataDoe publishes your Amazon data as a listing in Google's Analytics Hub, in two flavours: Integrated, which is the cleaned and joined schema, and Raw. You subscribe to the listing and get a linked, read-only dataset inside your own Google Cloud project. On the Base plan there is no monthly fee for the integration; queries draw on your AI Tokens, and Google bills the compute.
From there, the BigQuery connection in Claude points at a dataset that already holds Seller, Vendor and Ads tables. If you want the background on why a warehouse is worth having at all, read how to build an Amazon data warehouse in BigQuery.
Both routes are laid out side by side on the Claude Data for Amazon page.
Amazon's free Selling Partner plugin is in Claude's connector directory, and it gives a single third-party seller in the US a view of Seller Central sales, stock and traffic. Our guide to the plugin covers its scope. Whatever tool draws the chart, it can only chart what the connection reaches. With one US Seller Central account as the only source, these stay off the page:
DataDoe reads all three sources through Amazon's official APIs, joins them by SKU and ASIN with your costs across all Amazon marketplaces, and stores each day so earlier periods stay available. It fetches daily at 2:00 in the seller's local time, with some tables updated during the day. That is 130+ documented tables a dashboard could draw from. The plugin and DataDoe can sit on the same Claude account.
Anthropic's own advice is to say what you want to know, not only what to chart, and to name the table or app if you know it. Five requests written that way:
For the full walk-through of the first one, see how to build an Amazon sales dashboard in Claude Data. For how certified queries, live sharing and data stories work in general, see Claude Dashboards is now Claude Data.
On Claude Team plans, Whiteboards, Sheets and Data are all on by default. On Enterprise plans, Whiteboards and Sheets start switched off, and Claude Data switches on by default on 19 November 2026. Owners change any of this under Organization settings → Artifacts.
Live sharing deserves a careful read before an agency uses it. A dashboard shared this way runs on the owner's access, and viewers see the results of the queries the owner approved, even if they could never open the source themselves. Sharing Amazon data outside your own organisation is governed by Amazon's data rules and by your client agreement, and no buyer personal data belongs in a dashboard. More on multi-account work is on our page for agencies.
Dashboards, sheets and boards only read. DataDoe can also change things on Amazon through Actions, and the rule there is unchanged: Actions stay off until you enable them, most AI clients ask before a change, and every change is logged.
Yes. The beta that opened on 8 October 2026 under the name Claude Dashboards has been called Claude Data since 10 November, the day it became generally available. Anthropic says the new name reflects that it now does more than dashboards.
All paid plans. Team has them on by default; Enterprise owners decide under Organization settings → Artifacts.
From DataDoe, by one of two routes. The first is the DataDoe connector you add to Claude, because Claude Data draws on connected apps. The second is BigQuery: DataDoe lists your Seller, Vendor and Ads tables there, you subscribe, and Claude Data reads that dataset like any other warehouse.
No. A dashboard runs read queries. Changes to Amazon happen only through Actions, which are off until you turn them on.
Not if you run one US Seller Central account and that is all you want to see. You need a data layer once the page should include ads, Vendor Central, profit per SKU, last year, other countries or several accounts. DataDoe is $97 a month with a 14-day trial.
DataDoe is an independent company with no affiliation to Anthropic or Amazon. Product details for Claude Data, Sheets and Whiteboards come from Anthropic's announcement of 10 November 2026 and its Help Center, and can change while parts of them are in beta.
.svg)
