The Scenario
You are a sales enablement manager at a B2B SaaS company. The quarter just closed and someone in leadership wants an answer to the same question every post-mortem surfaces: what did customers actually say before we lost them?
You have 30 lost deals from Q1. Every one of them had at least one Gong-recorded demo. Someone wants to know whether price, timeline, or competitive pressure shows up most in the transcripts before you build the next training module.
The manual version of this project:
- Open Gong, search for calls tagged to lost deals, Q1 date range
- Click into each call one by one, read the transcript
- Copy text, paste into a Google Doc, highlight the objection words by hand
- Count occurrences per call, per objection type
- Build a summary table in Sheets from scratch
- Three hours later, you have a tally that covers twelve of the thirty calls because you ran out of morning.
The fast version is one prompt.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent inside your spreadsheet that reads Gong directly, so you do not have to open a single transcript by hand.
Open the SheetXAI sidebar and type:
Fetch call transcripts from Gong for calls that took place between January 1 and March 31 and paste each transcript's call ID, date, duration, and full text into separate rows in this sheet starting in row 2, with headers in row 1. Then for each transcript in column D, scan for the words "price," "budget," "competitor," and "timeline" and count occurrences of each in columns E, F, G, and H.
SheetXAI pulls the Gong call list for the date range, fetches each transcript, writes the metadata and text into the sheet, and then scans every row for your four objection signals. Every call, every count, in one pass.
What You Get
A populated sheet with one row per call:
- Column A — call ID
- Column B — call date
- Column C — duration in minutes
- Column D — full transcript text
- Columns E through H — occurrence counts for "price," "budget," "competitor," and "timeline"
The counts let you sort immediately. If "price" shows up 4 times per call on average but "competitor" shows up once, that is the finding. You did not have to read a single transcript to get there.
Want to go further? Tell SheetXAI to add a summary row at the bottom totaling each column, or to generate a brief paragraph at the top of the sheet describing the pattern across all 30 calls.
What If the Data Is Not Quite Ready
Real transcript pulls are rarely clean. SheetXAI handles edge cases in the same prompt.
When the transcript text is missing for some calls
Some calls were not fully transcribed — Gong occasionally misses short calls or calls where audio quality was low.
Fetch transcripts from Gong for Q1 calls. If a call's transcript is missing or blank, write "NO TRANSCRIPT" in column D and skip columns E through H for that row. Write all other rows normally.
When you want to scope to calls with specific rep or deal characteristics
You do not want all 30 lost deals — you want only the ones where a specific rep was on the call, or where the deal was above a certain size.
Fetch Gong transcripts for Q1 lost deals where the deal value in Gong is above $20,000. Write call ID, date, duration, and transcript text into columns A through D. Then count occurrences of "price," "budget," "competitor," and "timeline" in columns E through H. Skip calls with missing transcripts.
When you want keyword counts grouped by call stage
Not all objections land in the same part of a call. You want to know whether "price" shows up more in the first half or second half of transcripts.
For each transcript in column D, split the text roughly in half. Count occurrences of "price" and "competitor" in the first half versus the second half, and write those four counts into columns E through H labeled "price-early," "price-late," "competitor-early," "competitor-late."
When you want the full picture: pull, clean, count, and summarize in one go
No pre-built sheet, just a blank tab and a question from leadership due in an hour.
Fetch all Gong call transcripts from January 1 to March 31. Write call ID, date, rep name, duration, and full transcript into columns A through E. Skip any call with a missing transcript and write "NO TRANSCRIPT" in column E. Then scan column E for "price," "budget," "competitor," and "timeline" and write occurrence counts into columns F through I. Finally, write a 3-sentence summary in cell A1 describing which objection type appeared most frequently and in how many calls.
The pattern: instead of reading transcripts yourself and then counting, you ask for both in one prompt. SheetXAI does the reading, the extraction, and the synthesis.
Try It
Get the 7-day free trial of SheetXAI and open a blank sheet, then ask it to pull your last quarter's Gong transcripts and scan for whatever language you care about. The Gong integration is included in every SheetXAI plan. For related workflows, see how to pull call metadata for a pipeline review or the Gong in Google Sheets overview.