Grouped by what the server gives the model rather than by vendor. A sales team rarely needs more than four of these at once. The question to ask before adding one: which prompt is currently failing because this data is missing?
Contact and company search, email and mobile enrichment, and deep research on a named prospect. The research tool returns a structured brief rather than a paragraph, so the model can quote specifics instead of paraphrasing.
Read and write contacts, companies, deals, and engagements. Useful for pulling a deal's full activity history into a call-prep prompt.
SOQL queries, record CRUD, and metadata reads. Give the model a read-only profile first and widen scope once you trust the prompts.
Thread search, message reads, draft creation, and calendar reads. Draft creation matters more than send: a rep reviews before anything leaves.
Channel and thread search, message reads, and posting. Scope it to deal channels rather than the whole workspace.
File search and content reads across Docs, Sheets, and Slides. Point it at a shared drive that holds proposals, security answers, and pricing sheets.
Page and database search plus page creation. Common home for battlecards, ICP definitions, and objection libraries.
Issue search and status reads. Ask whether the feature a prospect asked about in March actually shipped.
Read-only SQL against your product database or a replica. The single highest-signal source for PLG expansion motions.
Pulls a URL and returns readable text. The plainest server on this list and one of the most used.
Scoped read and write access to a directory you name. Keep the allowlist narrow.
Read and append rows. Still the operating system of most sales teams whatever the CRM vendor says.
Take the last ten prompts you ran in Claude for work. Mark the ones where you pasted something in first. Whatever you pasted is the server you need. In most sales teams the pattern is boring and consistent: CRM history and a company or person you do not have data on.
That second one is the harder half. CRM servers are easy because the data is already yours. Prospect data is not: it has to come from somewhere, which is why a data server sits at the top of this list rather than a nicer looking productivity integration. The trade-off between assembling that layer yourself and renting it is laid out in this build-versus-buy breakdown of the AI SDR stack.
A shorter, opinionated cut of this list, with the reasoning for each pick, is on the Salesgear blog: 7 MCP servers every sales team should connect.