{"rewrite":{"id":"r_4785d43459735fb0921b7532","clusterId":"c_0a70f15634ce178a591e0e0a","slug":"anthropic-ties-claude-sales-agent-to-faster-close-rate","model":"deepseek-v4-1-flash","headline":"Anthropic Ties Claude Sales Agent to Faster Close Rate","summary":"Anthropic's sales team built a purchasing agent on Claude Managed Agents to handle tens of thousands of monthly inbound inquiries that once took days to answer. The first agent was built by one engineer in a few weeks, and the team reached version 7 about a week into internal testing. Leads handed off by the agent are more than twice as likely to reach a sales meeting, with time to close shorter by about 5 days.","whyItMatters":"The reported results give a concrete internal benchmark for agent tooling built by the same company that sells it, which matters to any team weighing how much of inbound sales work to hand to an AI agent.","webCardHtml":"\u003cp\u003eAnthropic\u0026#39;s sales development leader, Carl Johnson, says the team\u0026#39;s purchasing agent runs thousands of conversations a day and answers the most common questions before passing qualified leads to staff. The agent is opt-in: people submitting an inquiry choose between the agent and a sales representative. Ownership of the code stayed with engineering, while sales and content leads reviewed and edited system prompts in the console, with changes tested before reaching customers. All agent changes are versioned and can be reverted, and the team spent its time on prompts, tools, and the knowledge base rather than on hosting.\u003c/p\u003e","blueskyPost":"Anthropic's purchasing agent reached version 7 about a week into internal testing. The close-rate gain came from one engineer's few-week build, not a dedicated ML team.","twitterPost":"Anthropic's purchasing agent hit version 7 a week into testing. One engineer built it in a few weeks.","threadsPost":"Anthropic's purchasing agent reached version 7 about a week into internal testing, built by one engineer in a few weeks. The close-rate gain came from a small internal build, not a dedicated ML team.","newsletterBlurb":"Anthropic's sales team used Claude Managed Agents to build a purchasing agent for its inbound pipeline, which handles thousands of conversations a day and is opt-in for customers. The first agent was built by one engineer in a few weeks, and the team reached version 7 about a week into internal testing. Leads handed off by the agent were more than twice as likely to reach a sales meeting, with close time roughly 5 days shorter than leads from the standard form.","attributionJson":"[{\"source\":\"GIGAZINE\",\"url\":\"https://gigazine.net/news/20261001-inbound-claude-managed-agents/\",\"title\":\"Anthropic's sales team doubled its close rate with Claude's automated replies\"}]","lintFlagsJson":null,"lintHits":0,"costUsd":0,"inputTokens":4561,"outputTokens":573,"status":"published","repairAttempts":0,"nextRepairAt":null,"factsAttemptedAt":1791244334,"createdAt":"2026-10-05T23:50:30.000Z","publishedAt":"2026-10-05T23:51:10.000Z","updatedAt":"2026-10-05T23:51:10.000Z"},"cluster":{"id":"c_0a70f15634ce178a591e0e0a","canonicalTitle":"Anthropicの営業チームはClaudeの自動返信で成約件数を2.5倍にした","representativeArticleId":"a_721d8ca372f176cf336c16f3","sourceCount":1,"writtenSourceCount":1,"writeAttempts":0,"isSolo":true,"entitiesJson":"{\"anime_titles\":[],\"manga_titles\":[],\"work_titles\":[],\"studios\":[],\"people\":[],\"type\":\"business\",\"domain\":\"industry\",\"is_roundup\":false}","contentType":"news","status":"published","firstSeenAt":"2026-10-01T02:40:00.000Z","lastSeenAt":"2026-10-01T02:40:00.000Z","updatedAt":"2026-10-05T23:51:06.000Z"},"attribution":[{"source":"GIGAZINE","url":"https://gigazine.net/news/20261001-inbound-claude-managed-agents/","title":"Anthropicの営業チームはClaudeの自動返信で成約件数を2.5倍にした"}],"entities":{"anime_titles":[],"manga_titles":[],"work_titles":[],"studios":[],"people":[],"type":"business","domain":"industry","is_roundup":false},"keyFacts":null}