Dashboards · The data behind it
A dashboard is only as current as the data under it.
Staying current
New data in, updated board out
Boards refresh when data does
When a pipeline run finishes, boards built on that data refresh their draft automatically.
Publish new data automatically
Turn on auto-publish and new data goes live to readers as soon as it syncs.
Refresh schedule by plan
How often data refreshes depends on the plan: by hand, weekly, daily, every 3 hours, or event-driven.
Frequency depends on the plan
Safe on schema changes
If the columns behind a board change, the sync is held and the creator is told, instead of the board breaking.
More than one table
One board across your tables
Each page of a board can read a different table, so sales and targets, or orders and staffing, sit on one board.
Several tables on one board
Each page can read a different table, so one board can cover several tables.
Change data
Point any saved board at a different dataset; Mammoth proposes the field mapping, previews your real numbers, and refuses a poor match.
Sources panel
See each dataset a board reads, with links to preview rows, open the view or change data.
See which boards a dataset feeds
A view's links menu lists every board built on it, and deleting data first shows the boards it affects.
Getting time right
Dates and balances, handled
Trends by hour, day, week or longer
Trends run from years down to days, and short ranges can show hourly points.
Balances are not added up over time
Balances like cash, headcount or inventory on a snapshot table show the latest period instead of being summed across months.
Suspended data is left out
Rows in a suspended batch don't count on boards, in Q&A or in exports.
Limits
Limits worth knowing
Better to know before you choose than after.
- Readers see new data when you publish
- The board's draft updates when the data does; readers see it when you publish, or straight away with auto-publish on. How often data refreshes depends on the plan. A schema change pauses the sync.
- Auto-publish moves data, not design
- Off by default. Publishes data refreshes only; layout and chat edits still wait for Publish.
- How often depends on the plan
- Free refreshes by hand, Starter weekly, Team daily, Pro every 3 hours, Enterprise event-driven.
- Up to five tables, one per page
- Up to 5 tables, all from the board's own project. A chart still reads one table, and joins happen in the data editor or pipeline.
- Change data stays in the project
- Only datasets in the board's own project. It replaces the sources rather than adding to them. Chat won't do it without the review.
- Hourly views on short ranges only
- An hourly view is offered only on short ranges over timestamp columns.
- Boards that read several tables
- They can only be copied within their own project for now, and can't yet be exported to Power BI or Tableau, saved as a template, or applied to other data. Boards that read one table have none of these limits.
- Joining tables happens in the pipeline
- A single chart reads one table. To chart two tables together, join them in the data editor first; the join is recorded and re-runs with every refresh.
Checked against the product on 2026-10-23. Every dashboard capability, with its plans and limits, is also published as machine-readable JSON.
Questions people ask first
How often does a dashboard refresh?
What happens if my source changes its columns?
Can one dashboard use several tables?
Can I reuse a dashboard on new data?
Build a board that updates itself.
Connect a source once; the pipeline and the dashboard keep up from there.
- ✓ 14-day Pro trial
- ✓ No credit card
- ✓ Viewers always free