SpyBara
Go Premium

use-cases/consolidate-spreadsheets.md 2026-07-27 18:59 UTC to 2026-07-28 23:01 UTC

21 added, 0 removed.

2026
Fri 31 22:02 Thu 30 23:58 Wed 29 22:01 Tue 28 23:01 Mon 27 18:59 Fri 24 15:00 Thu 23 21:57 Wed 22 20:02 Tue 21 22:02 Mon 20 23:01 Fri 17 22:57 Thu 16 20:57 Wed 15 19:58 Tue 14 17:03 Wed 8 02:01 Mon 6 22:58

name: Consolidate spreadsheets tagline: Join spreadsheet exports into one refreshable workbook. summary: Give ChatGPT spreadsheet exports, join keys, targets, segment definitions, and reporting rules, then ask it to create a consolidated workbook with cleaned joins, charts, insights, assumptions, refresh instructions, and mismatch review. skills:

  • token: $spreadsheets description: Inspect exports, clean joins, build formulas, and validate the resulting workbook.

  • token: google-drive url: https://github.com/openai/plugins/tree/main/plugins/google-drive description: Read approved spreadsheet sources and reporting templates from Drive.

  • token: $documents description: Add clear assumptions, refresh instructions, and a review summary. bestFor:

  • Reporting workflows that combine multiple exports into one updateable workbook.

  • Pipeline, account, operations, or performance analysis with explicit join keys and targets.

  • Teams that need mismatched records and refresh assumptions isolated for review. starterPrompt: title: Consolidate these exports body: >- Consolidate the attached spreadsheet exports into one updateable workbook.

    Join on [record or account key]. Clean duplicate records, calculate [metrics or views], compare [actuals or pipeline] with [target], and create a dashboard with charts and plain-English insights.

    Include:

    • assumptions and refresh instructions

    • a review section for mismatched keys or records that did not join cleanly

    • a short change log

    Do not invent join keys or silently merge conflicting records. Preserve the original files and return the consolidated workbook for review. suggestedEffort: high relatedLinks:

  • label: "OpenAI Academy: Everyday work" url: https://openai.com/academy/how-to-use-codex-for-everyday-work/

  • label: Spreadsheet skills url: /codex/build-skills

For the complete documentation index, see llms.txt. Markdown versions of documentation pages are available by appending .md to the page URL.

Establish the join and reporting rules

Consolidation is safest when the key, field definitions, target calculations, and duplicate rules are explicit. Give ChatGPT the source exports and ask it to surface records that do not join cleanly instead of guessing.

  1. Attach the exports, mapping notes, targets, and reporting requirements.
  2. Confirm the join key, duplicate rule, period definitions, and required output views.
  3. Run the starter prompt and request an editable workbook with a review section.
  4. Inspect formulas, joins, mismatches, assumptions, and refresh instructions.
  5. Re-run the checks after any change to the source files or reporting rules.

Keep the original exports unchanged. A useful consolidated workbook makes it easy to tell which rows joined, which were excluded, and which assumptions a future refresh depends on.

Validate the refresh path

After the first workbook is ready, ask ChatGPT to simulate the next refresh and report which steps are repeatable and which still need a human decision.

Prompt:

Test the refresh process for this consolidated workbook.

Report:

- inputs that can be replaced without changing the workflow
- formulas, joins, and charts that update correctly
- duplicate or mismatched keys
- assumptions that are not documented
- rows that need manual review
- steps that should become a reusable skill or automation

Do not overwrite the source exports or change business definitions.