The Scenario
You're a UX researcher. You ran 15 user interviews over the past two weeks — all recorded in tl;dv. You need a single Excel workbook where each row is one interview: meeting name, date, duration, the AI-generated highlight summary, and the first 500 characters of transcript. Your stakeholder wants to scan the whole study at a glance before the readout next Tuesday.
The recordings exist. The data exists. What doesn't exist is the combined view.
You could pull the metadata for each meeting, then go back and pull highlights, then go back again for transcript previews. Three separate passes across 15 rows. More likely, you start doing it manually and realize twenty minutes in that you're only on interview four and the readout is in three days.
The bad version:
- You open tl;dv and start clicking through each interview to copy the metadata into your workbook.
- You finish the metadata for all 15 calls, then start the second pass for highlights — copy, switch tabs, paste, repeat.
- By the time you get to the transcript previews you're on your third hour of data assembly. The actual synthesis work hasn't started.
Nobody hired you to copy-paste tl;dv data into a spreadsheet. The insight is what you're being paid for. The data pipeline is overhead.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent that lives inside your Excel workbook. It reads your workbook, connects to tl;dv, and can pull metadata, highlights, and transcript previews for multiple meetings in a single pass — one row per interview, all columns filled at once.
Open your Excel workbook and type into the SheetXAI sidebar:
For each meeting ID in column A, fetch the tl;dv meeting details, highlights, and first 10 transcript lines — then populate columns B through F with meeting name, date, duration, highlight summary, and transcript preview (first 500 characters of transcript text).
What You Get
- Column A: Your existing meeting IDs (already there)
- Column B: Meeting name from tl;dv
- Column C: Interview date
- Column D: Duration in minutes
- Column E: AI-generated highlight summary
- Column F: First 500 characters of the transcript
- One row per interview, ready for your readout slide
What If the Data Is Not Quite Ready
I don't have the meeting IDs yet — I need to discover the right calls first
Fetch all tl;dv meetings from the last 30 days where the meeting name contains "user interview" and write their meeting IDs into column A — one row per match.
Some interviews don't have highlights yet because processing is still running
For each meeting ID in column A, fetch the tl;dv details, highlights, and transcript preview — write all available fields into columns B through F, and for any meeting where highlights are still processing write "Pending" in column E.
I want to filter by organizer — only my interviews, not the whole team's
For each meeting ID in column A, fetch the tl;dv meeting details and check the organizer field — if the organizer email matches mine, pull the highlights and transcript preview and write them into columns B through F. If the organizer doesn't match, write "Different organizer" in column B and leave the rest blank.
Full research log: metadata + highlights + transcript preview + auto-categorization
For each meeting ID in column A, fetch the tl;dv details, highlights, and first 500 characters of transcript — write name, date, duration, highlights, and transcript preview into columns B through F. Then read the highlight text and add a "Theme" label in column G: use "Navigation" if the highlights mention finding or searching, "Frustration" if they mention confusion or errors, "Delight" if they mention positive reactions, or "Other" for everything else.
One pass: data pull, field mapping, and thematic coding all in a single prompt.
Try It
Get the 7-day free trial of SheetXAI and open any Excel workbook where you're planning a research readout, then ask it to pull tl;dv metadata, highlights, and transcript previews into a combined row for each interview. You can also export your full meeting registry with the meeting registry spoke, or see the full tl;dv overview.