Quote discount approval summaries
When a rep requests a discount above policy, an agent prepares a one-page summary with deal context, margin impact, and past pricing for similar customers so the approver can decide quickly.
The use case
Discount requests often arrive as a short message: can I give this customer 20 percent? The approver has to dig through the CRM, the price list, and past deals to understand whether it makes sense, so decisions are slow or made without context. When a rep submits a discount request, the agent gathers the deal details, customer history, product mix, and list price, estimates margin impact using the cost data you provide, and finds recent deals of similar size and type to show what discounts were given. It notes anything unusual, such as a long contract term or a competitor mentioned in notes. The approver reads the summary and approves, counters, or declines. The agent does not change quote prices, approve discounts, or send quotes to customers.
Example workflow
- A rep submits a discount request from the CRM deal record or a short form with the requested price and reason.
- The agent pulls deal details, products, quantities, and customer history from HubSpot or Salesforce.
- It calculates the discount and estimated margin using the price list and cost data stored in Google Sheets.
- It finds recent comparable deals and lists the discounts given and their outcomes.
- A one-page summary is sent to the approver in Slack, who approves, counters, or declines in the CRM.
- The rep updates the quote and sends it to the customer after approval.
Tools
- HubSpot
- Salesforce
- Google Sheets
- Slack
- Claude
- PandaDoc
Risks and guardrails
- Margin estimates are based on outdated cost data — guardrail: the summary shows the date of the cost data, and the agent flags costs older than a set age.
- Comparable deals are not truly comparable — guardrail: the summary lists each comparison deal so the approver can judge relevance.
- Discounts are applied without approval — guardrail: the agent cannot edit quotes, and approval is recorded in the CRM by a person.
- Cost and margin data are visible to people who should not see them — guardrail: summaries go only to approvers in a private channel.
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:
- HubSpot MCP server (official page): developers.hubspot.com/ai-tools/mcp (checked 2026-11-09)
- Salesforce MCP servers (official page): www.salesforce.com/agentforce/mcp-support (checked 2026-11-09)
- Slack MCP server (official page): docs.slack.dev/ai/slack-mcp-server (checked 2026-11-09)
- Claude (official page): claude.ai (checked 2026-11-09)
- PandaDoc MCP server (official page): developers.pandadoc.com/docs/use-pandadoc-mcp-server (checked 2026-11-09)