name: Diagnose a stalled deal tagline: Find the real blocker and the next customer-facing move. summary: Give ChatGPT stage history, closed activities, call transcripts, emails, deal threads, security or procurement notes, and account context, then ask it to explain the blocker, prior attempts, escalation path, and next action. skills:
-
token: $spreadsheets description: Trace stage history, activity, and timing signals from CRM exports.
-
token: google-drive url: https://github.com/openai/plugins/tree/main/plugins/google-drive description: Read account context, security or procurement notes, and prior deal plans.
-
token: slack url: https://github.com/openai/plugins/tree/main/plugins/slack description: Review deal threads and internal escalation context.
-
token: gmail url: https://github.com/openai/plugins/tree/main/plugins/gmail description: Check customer and internal email history that explains the stall. bestFor:
-
Deals that have stopped moving despite recent activity.
-
Sales teams separating a real customer blocker from internal process or follow-up gaps.
-
Escalation reviews that need a clear next customer-facing move and owner. starterPrompt: title: Diagnose the stalled deal body: >- Diagnose why [deal or account] is stalled.
Use the opportunity stage history, closed activities, call transcripts, email threads, deal threads, security, legal, or procurement notes, and account context I provide.
Return:
-
the most likely blocker and supporting evidence
-
prior attempts and what changed
-
customer, technical, legal, or procurement dependencies
-
escalation path and owners
-
the next customer-facing move
-
missing context and risks
Separate sourced facts from inference. Draft follow-up language only; do not send it or update CRM records. suggestedEffort: medium relatedLinks:
-
-
label: "OpenAI Academy: Sales teams" url: https://openai.com/academy/codex-for-work/how-sales-teams-use-codex/
-
label: Turn meetings into follow-ups url: /codex/use-cases/zoom-meeting-follow-ups
Trace the stall across the full deal history
The visible stage is rarely the whole blocker. Give ChatGPT stage history, closed activities, calls, emails, deal threads, security or procurement notes, and account context so it can distinguish customer, technical, legal, and internal process issues.
- Define the deal, time window, and decision the diagnosis should support.
- Attach stage history, activity, transcripts, email, threads, security or procurement notes, and account context.
- Ask ChatGPT to build a timeline before naming a root cause.
- Run the starter prompt and review evidence, prior attempts, dependencies, escalation path, and next move.
- Have the deal owner approve any customer-facing follow-up before sending it.
Keep the diagnosis factual and time-bounded. If the blocker is uncertain, list the smallest customer or internal check that would distinguish the competing explanations.
Turn diagnosis into a recovery plan
After the blocker is agreed, ask ChatGPT to create a narrow recovery plan with owners, dependencies, and a proof point for each step.