> For the complete documentation index, see [llms.txt](https://guides.tability.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://guides.tability.io/docs/become-a-tability-power-user/ai-mode-how-tos/export-review-spreadsheet.md).

# Prepare a team review spreadsheet

Build a detailed review table from Tability data and download a CSV for Excel or Google Sheets.

Bring the detail behind a team review into a spreadsheet you can sort, filter, and annotate. AI Mode can gather the relevant records and create a downloadable CSV with source links.

**For:** Managers · **Where:** AI Mode inside Tability

## Watch the walkthrough

{% embed url="<https://318707989-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMULqjSmZyGczPy2I6FGH%2Fuploads%2Fgit-blob-669e4f5d7bbce6ea2f62ca16366437c010c77539%2Ftability-ai-mode-spreadsheet.mp4?alt=media>" %}
Export active key results, owners, progress, confidence, and check-in dates as CSV. 30-second demo, no audio; fictional workspace data.
{% endembed %}

## 1. Define the rows and columns

Open **AI Mode** and identify the team or plans to review. For a goal review, use one row per key result:

```
Create a downloadable CSV for the active key results in [team or plan links]. Use one row per key result with these columns: Plan, Key result, Owner, Current value, Target, Unit, Progress, Latest check-in date, Confidence, Evidence of blockers, Proposed next action, and Source URL.

Use exact record titles and verified values. Keep missing numbers empty and describe missing evidence in text columns. Distinguish proposed actions from existing commitments. Include all accessible records in scope and tell me if coverage is incomplete. Show the CSV in the artifact panel with a short summary in chat.
```

Add a reporting cutoff if the review covers a past period, and ask for historical observations rather than today's values. If the dataset is large, narrow it to named plans or split it into clearly labelled exports.

## 2. Inspect the table

Check the preview's row and column counts. Spot-check a few source links, dates, owners, and values, including records with missing check-ins. Use follow-ups to change the export:

```
Keep the same scope and add a column showing whether the latest check-in is more than seven days old. Put the stale records first. Preserve all source rows and URLs.
```

## 3. Download and use it in a spreadsheet

1. Select the final table in the artifact panel.
2. Click the download icon and choose **CSV (.csv)**. The download includes all rows in that artifact, including rows on other preview pages.
3. Import the file into Excel or Google Sheets. Preserve identifier columns as text and check how dates and decimal values were interpreted.
4. Add your own filters, formulas, annotations, or pivot tables in your spreadsheet app.

The export contains values, not formulas or a live connection to Tability. The preview is read-only, and changes in your spreadsheet do not update Tability. Request a fresh export for the next review. Keep different metric units separate when calculating totals or averages.

**Done when:** the file opens with the expected rows and columns, and reviewers can trace each row back to its source.

Explore the other [AI Mode how-tos](/docs/become-a-tability-power-user/ai-mode-how-tos.md).