Analyze Lost Deals with Claygent and OpenAI
The problem
Lost deal reasons in CRM are vague ("pricing", "timing"). Reps rarely document specifics.
Implementation steps
- 1
Export closed-lost deals from HubSpot (last 90 days) to Clay
- 2
Use Claygent to scrape deal notes and email threads
- 3
Send notes to OpenAI API with prompt: "Categorize loss reason: competitor, budget, no decision, feature gap, or other"
- 4
Aggregate results in Google Sheets with pivot table
- 5
Share monthly report with sales leadership
Template outline
The structure to recreate in your own workspace. Field names and thresholds are starting points.
Clay + OpenAI Workflow:
1. HubSpot integration: Pull closed-lost deals (last 90 days)
2. Claygent: Fetch activity timeline and notes
3. OpenAI GPT-4: Classify loss reason from notes
- Prompt: "Given these deal notes: {{notes}}, classify the loss reason as one of: Competitor, Budget, No Decision, Feature Gap, Other. Reply with one word."
4. Google Sheets: Aggregate by loss reason
5. Create pivot table for monthly reviewKeep going
More Research playbooks
- ResearchBeginner
Monitor Competitor Hiring to Spot Market Moves
Competitor intelligence is reactive. You learn about their new product after launch, not during the build.
ClayClaygentSlackEstimate: 2 hrs/week - ResearchIntermediate
Analyze Firmographics of Closed-Won Deals to Refine ICP
Your ICP doc was written years ago, and the company profile it describes may no longer match the deals you actually win.
ClayHubSpotClearbitEstimate: 4 hrs/week - ResearchIntermediate
Use Claygent to Research Target Account Tech Stacks
Sales needs to know if prospects use competing tools before the first call, but researching each account by hand is slow.
ClayClaygentBuiltWithEstimate: 7 hrs/week
Need help implementing a playbook?
Our team builds custom Clay workflows and AI automations for revenue teams.
Get it built for you