Resource
Executive Summary Prompt
Summary, Why It Matters, What Happens Next — the executive summary contract for readers who will never open the source.
Customer Call Summary Prompt
Action items only: every commitment from the call — owner, task, deadline — straight into the CRM, with quotes for the promises that matter.
Overview
After a customer call, the summary nobody reads is the long one; the artifact that matters is the commitment list. This setup uses the Action Items structure on support-conversation sources: every commitment made on the call as one bullet — owner when stated, task, deadline when stated — plus an Open Questions section for what's blocking. Important Quotes keeps the promises that matter in the customer's exact words, and the conversation guidance separates what the customer reported from what the agent concluded.
How to use this resource
-
Run per call transcript
One call, one commitment list — pasted into the CRM activity as-is.
-
Check the discussed-vs-committed bar
"A topic discussed is not an action item" — the rule that keeps wish lists out of the follow-up queue.
-
Keep promises verbatim
Important Quotes preserves the exact commitment — useful when "by Friday" gets remembered as "sometime next week".
Why This Works
- The action-item structure produces the artifact teams actually use after a call
- Owner — task — deadline formatting makes follow-ups assignable at a glance
- Reported-vs-concluded separation keeps the record factual
Best for
- Sales and CS teams logging calls at volume
- Follow-up workflows keyed on owners and deadlines
- Calls where the promise's wording will matter later
Not for
- Extracting call metadata into named fields for an API — that's extraction; this writes a readable record
- Full conversation records — the Support Case Summary preset keeps the narrative
Use cases
- Logging call follow-ups into the CRM in one consistent shape
- Capturing who promised what, in their words
- Separating customer-reported facts from agent conclusions
FAQ
Where do the promise quotes go if the output only has Action Items and Open Questions sections?
They stay inline inside your two sections, not in a separate quotes block. The output structure fixes exactly two headings — ## Action Items and ## Open Questions — while the QUOTE RULES tell the model to quote the most important statements exactly, in quotation marks, attributed to a speaker. So a promise like 'we'll refund by Friday' rides along its Action Items bullet, quoted verbatim rather than paraphrased.
How does the prompt handle a commitment where no owner or deadline was actually stated on the call?
It leaves them out instead of guessing, since the Action Items bullet format is 'owner (when stated) — task — deadline (when stated)' and the fidelity rules bar adding outside knowledge. A vague 'we'll look into it' becomes a task with no owner or date, not a fabricated name or deadline. The assistant only compresses wherever you paste the prompt; it can't recover facts the transcript never held.
Why does the summary split what the customer reported from what the agent concluded?
The SOURCE reading guidance treats those as different facts, so a symptom the customer described stays separate from the agent's diagnosis of it. That keeps a reported complaint from being logged in the CRM as a confirmed cause. It's a framing instruction inside the generated prompt, not a fact-check — the tool structures the record faithfully, but confirming the diagnosis is still your call.
More resources from Structured Summary Prompt
Resource
Structured Summary Prompt — the Anatomy
The blocks a reliable summary prompt needs: source guidance, a fixed section skeleton, length budgets, fidelity rules, and quote handling.
Resource
Stop AI Summaries from Making Things Up
The fidelity ladder: Balanced, High, Strict — and the six-rule battery that keeps a summary inside its source.
Resources that pair well
Resource
Markdown Output Prompt — Lock the Document Structure
The contract that stops AI documents from restructuring themselves: a pinned section skeleton, forced tables, and strict consistency rules.
Resource
Extract Data From Text with AI
Free text in, named fields out. The extraction prompt pattern that turns any unstructured text into consistent, parseable records.
Resource
Fix a Vague Prompt
'Make it good', 'be detailed', 'keep it interesting' — vague prompts get vague output. The fix is mechanical: replace every fuzzy word with a checkable instruction.
Related tools
Tool
Structured Summary Prompt
Build summary prompts with fixed sections, length caps, and no-invention fidelity rules.
Workflows that use this resource
Workflow
AI Customer Support Workflow
Run inbound support the same way every time — triage and route the ticket, pull the details that matter, draft a reply in a consistent voice, and log the resolution for the record.