Claude Code prepares Apollo and Attio outreach drafts
Braadbaart reports Claude Code prescreening Apollo companies, creating Attio records and company PDFs, then placing outreach drafts in Gmail for his review and manual send.
Original prompt
Shared by the source author. See source notes for context.
/lead-source
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source author
- Jonas Braadbaart
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- Braadbaart reports Claude Code prescreening Apollo companies, creating Attio records and company PDFs, then placing outreach drafts in Gmail for his review and manual send.
- Human involvement
- Braadbaart manually selects targets and reviews and sends drafts. The source does not independently establish closed customers, and screening percentages are personal reports.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Filter and export the intended company population in Apollo.
- Prescreen and rank companies against the agency’s context and evidence.
- Select targets and source decision-maker contacts through API.
Jonas Braadbaart builds a sales workflow for his AI agency using Claude Code, Apollo, Perplexity, Attio and Gmail. He manually filters and downloads company lists in Apollo, then runs a prescreening skill that researches fit against his service-heavy European business criteria.
After ranking the screened companies, he chooses an outreach list. A lead-sourcing skill retrieves likely decision-maker contacts through Apollo API calls, creates Attio CRM records and prepares company-specific PDF reports describing possible operational bottlenecks. A custom Gmail MCP places the draft email in the sales inbox with the report attached; a LinkedIn message draft is also prepared.
Braadbaart checks and edits the report and email before sending. Attio's mailbox connection then records actual interactions, including who should not be contacted again. The source gives personal screening percentages and time estimates, but no independently verified closed-customer result. A commenter explicitly asks whether the system has booked customers, so preparing a researched outreach set should not be presented as a proven sales win. The useful chain keeps company screening, contact sourcing, CRM logging, report creation and draft review visible as distinct stages, with a dedicated sales account and a human final send.
Prompt
Original prompt
Source: Blog · Jonas Braadbaart
Frequently asked questions
What remains manual before research?
The author filters company lists in Apollo’s interface and downloads the starting population.
Where does outreach wait for review?
The email draft is created directly in the dedicated Gmail sales account with the company PDF attached.
Did the account prove new customers?
No. It describes the working preparation pipeline and personal estimates, not independently verified closed-customer results.
More details
Remaining steps
- Create Attio records, company PDFs and Gmail outreach drafts.
- Review and edit the artifacts, then send manually.
About this account
Claude Code researches companies, sources contacts and writes CRM records, PDFs and message drafts through connected tools while the sender retains the final contact decision.
Firsthand account by Jonas Braadbaart; original publication date verified against Primary Substack Apr 13, 2026. Outcomes are the author’s reports, not independently reproduced. Original custom lead-sourcing command from the working sales system; the full skill and prospect inputs are not disclosed.
Credit: Jonas Braadbaart
Source published
Added to Agents In Use
Source checked
Tools and topics
Claude Code · Apollo API · Perplexity · Attio MCP · Custom Gmail MCP · Company PDF reports
Prescreened outreach · CRM provenance · Manual send
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.