Sales objection library from call notes
An agent reads call notes and transcripts to collect the objections prospects raise, groups them, and drafts suggested responses based on what worked before, for the sales lead to review and approve.
The use case
Prospects raise the same objections again and again: too expensive, we already use another tool, now is not the right time, we need approval from finance. Experienced reps know good answers, but newer reps learn them slowly, and the answers live in people's heads. The agent reads recent call notes and transcripts, extracts the objections raised, and groups them by theme. For each theme, it finds examples where a rep's response led to a next step and drafts a suggested response based on those examples and the company's approved positioning. It marks any response that makes product or pricing claims for checking. The sales lead reviews, edits, and approves each entry before it goes into the team's playbook. The agent does not invent claims, testimonials, or results.
Example workflow
- Each month, call notes and transcripts are exported from Gong, Fathom, or the CRM.
- The agent extracts objections and groups them by theme, counting how often each appears.
- For each theme, it finds examples where the rep's response led to a next step, such as a follow-up meeting.
- It drafts a suggested response using those examples and approved positioning documents, marking any factual claims.
- The sales lead reviews, edits, and approves entries in Notion, removing anything unsupported.
- Approved entries are added to the sales playbook and reviewed again each quarter.
Tools
- Gong
- Fathom
- HubSpot
- Notion
- Claude
Risks and guardrails
- Responses include claims the product cannot support — guardrail: factual claims are marked, and the sales lead checks them before approval.
- Prospect names or confidential details appear in the playbook — guardrail: examples are anonymized and stripped of company names.
- A next step after one response is mistaken for proof it works — guardrail: entries note how many examples support each suggestion.
- Call recordings are used without proper notice — guardrail: only calls recorded under the company's notice and consent practices are used.
Related recipe
Inbound Lead Qualification and Reply Drafts: An agent scores each inbound lead against your written ICP, checks the company's public website, and drafts a reply with a suggested next step that a rep reviews and sends; nothing goes out automatically.
Want it set up for you? Request a quote.
Sources
Written by Agents Lookup. Credits and official pages for the tools this use case names:
- Gong MCP server (official page): help.gong.io/docs/about-gong-mcp-server (checked 2026-11-09)
- Fathom (official page): www.fathom.ai (checked 2026-11-09)
- HubSpot MCP server (official page): developers.hubspot.com/ai-tools/mcp (checked 2026-11-09)
- Notion MCP server (official page): developers.notion.com/guides/mcp/overview (checked 2026-11-09)
- Claude (official page): claude.ai (checked 2026-11-09)