Customer Support & Ops

Verified against ChatGPT · 2026-08-13

Log a support call into the CRM in the exact fields your team actually uses, without adding editorial the next agent will trust as fact

Turns raw call notes or a transcript into a CRM-ready call summary formatted to your actual field structure, with next steps assigned to a named owner and sensitive details handled per your policy.

ChatGPT (GPT-5.1)5 fillable variables

The prompt

Ready to copy — highlighted parts are example details you can swap.

Turn this call into a CRM log entry, formatted to the fields we actually use, not a general-purpose call summary.

CALL TRANSCRIPT OR NOTES
Customer called about a billing discrepancy, mentioned in passing they're evaluating a competitor, agreed to a callback next Tuesday once the discrepancy is confirmed resolved.

CALL PURPOSE
Billing discrepancy follow-up.

NEXT STEPS OWNER
Account manager (Dana) owns the Tuesday callback; billing team owns confirming the discrepancy correction by Monday.

CRM FIELD CONSTRAINTS
Fields are: Summary (2 sentences max), Next Steps (bulleted, owner + date required), Risk Flags (short, only if applicable).

SENSITIVE INFO HANDLING
Competitive-evaluation mentions go in a separate restricted 'Account Risk' note visible only to the account manager, not the main shared summary.

Write the summary strictly in the given CRM field structure, in the length each field actually allows — do not pad a short field with detail that belongs in a longer one, and do not compress something important into a field too short to hold it; flag it as a mismatch instead of silently cutting it. Report only what was actually said or agreed on the call — do not add interpretation, sentiment judgment ("customer seemed frustrated") unless it's directly relevant to a next step, or a next-step that wasn't actually discussed on the call. A CRM log is read later by people who weren't on the call and will treat every line as fact, so anything uncertain or inferred must be marked as such rather than blended in with what was confirmed.

Assign every next step to the specific owner named, with what "done" looks like for that step — a next step with no owner or completion criteria tends to sit untouched until the customer follows up asking why nothing happened.

Apply the sensitive info handling rule exactly as given — if something discussed on the call shouldn't go into a permanent, broadly-visible CRM record, note that it was discussed and where it's actually documented instead (a separate restricted note, a ticket), rather than either omitting it entirely or logging it somewhere the policy says it shouldn't be.

OUTPUT FORMAT
The CRM entry itself, field by field as specified, plus one line at the end flagging anything that didn't fit cleanly into the given field structure.

Customize

Optional — swap in your own details for the highlighted parts above.

Why this works

Forcing the summary into the exact given field lengths, rather than a free-form recap, matters because GPT-5.1's default call-summary shape is a well-organized but variable-length narrative, and asked to just "summarize this call" it will produce something that reads well but doesn't respect a CRM's actual field constraints — the instruction to flag a mismatch rather than silently compress or pad forces an honest signal back to the person logging it about where the real content doesn't fit the system, instead of quietly losing information to fit a box. The instruction against adding interpretation or unconfirmed sentiment addresses a specific risk of CRM records: they are read later, out of context, by people who take every line as an established fact rather than a live impression, so a model's natural tendency to add a smoothing sentence like "customer seemed satisfied with the resolution" becomes a permanent, uncritically-trusted data point even if it was actually just a plausible-sounding inference from tone, not something the customer said. Routing the sensitive detail (the competitive-evaluation mention) to the specified separate restricted note rather than either dropping it or leaving it in the shared record is the mechanism that actually respects the stated policy — a model not given this explicit routing instruction will default to including everything discussed in the main summary because omitting a detail feels like information loss, without knowing that this particular detail has a policy-mandated home elsewhere.

What you get back

Summary: Customer called regarding a billing discrepancy on their most recent invoice; discrepancy correction is pending confirmation, with a follow-up callback scheduled. Next Steps: Billing team to confirm the discrepancy correction by Monday (owner: billing team). Dana (account manager) to call back Tuesday to confirm resolution with the customer. Risk Flags: none in this record — see restricted Account Risk note for a related item raised during the call. Mismatch flag: the customer's passing mention of evaluating a competitor doesn't fit the shared Summary field per policy and has been routed to the restricted note instead.

Verified against

ChatGPT GPT-5.1 · 2026-08-13

Changelog

  • 2026-08-13 Initial publish, verified against ChatGPT GPT-5.1.

Need this built into your business?

If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.

EXPLORE WHAT SCULT BUILDS
All Customer Support & Ops prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY