Refresh or Replace a Dataset to Update Your Charts

When you connect a data source, ChartBrick takes a snapshot of it and stores that snapshot. Every chart and dashboard reads from the snapshot, not from the live source. That keeps charts fast and consistent, but it means new rows in your spreadsheet or database won't appear until you update the dataset.

There are two ways to update, depending on the dataset:

• Refresh — re-pulls data from a connected source (Google Sheets, Airtable, Notion, MySQL, PostgreSQL, Stackby, API URL, or a CSV/JSON pulled from a URL).
• Replace — uploads a new file over an uploaded dataset (CSV, JSON, or Excel files).

In the sidebar, click New to open the datasets page, then click the dataset you want to update. The header shows its source, row count, column count, and when it was last refreshed (for example, "Refreshed 2 hours ago").

The action button on the top right changes based on the dataset type. Connected sources and uploaded files both show an Update data button; the button does the right thing for that dataset.

Note: Viewers can't refresh or replace data. You need an Editor or Admin role on the workspace.

For a dataset backed by a live source, click Update data. ChartBrick queues the refresh and runs it in the background, so the button shows a loading state while the work happens. Your manual click jumps ahead of any scheduled refreshes.

A few things to know:

• If a refresh is already running for that dataset, your click joins the one in progress instead of starting a second one.
• Static datasets can't be refreshed: Excel files, sample datasets, and CSV/JSON you uploaded as a file or pasted as text. For those, replace the file instead (next step).
• If a refresh fails, the button turns red and shows the error on hover — fix your source or credentials and try again.

For a dataset you created by uploading a file, click Update data and pick the new file. The file's type must match the dataset:

• CSV datasets accept .csv
• JSON datasets accept .json
• Excel datasets accept .xlsx or .xls

The maximum upload size is 100 MB. If the new file renames or drops columns, ChartBrick re-maps your charts to the new data automatically and shows a Charts updated notice listing what changed. If everything still lines up, you'll see Data replaced instead.

Open a dashboard, click the More options (⋮) button in the header, and choose Refresh all datasets. ChartBrick finds every distinct dataset used by the dashboard's charts and queues each one to refresh in the background. A toast reports the outcome, such as "3/4 datasets refreshing in the background."

The same rules apply per dataset: static datasets (Excel, samples, uploaded files) can't be refreshed and are called out in the toast as needing a manual replace. Only refreshable sources are queued this way — uploaded files still have to be replaced one by one from their dataset page.

Refreshes run in the background, so give them a moment to finish. The dataset header updates the "Refreshed" time and the Data preview table once the new snapshot lands. Open any chart that uses the dataset to confirm it reflects the latest numbers.

If the dataset shows a This dataset is missing rows warning, the source returned more rows than were imported — the snapshot is partial until the next clean refresh clears it.