We would like to have separate databases for each client. They will have access to dashboards we create with their respective data but we would like to have some dashboards for internal use that combine the data from multiple clients. Merging collections in different db's would accomplish this.

On large collections, loading filters potential values is very resources consuming. It would be great to not load all the data when displaying the filters.
A button to toggle this behaviour would be great.

Do not load graph before validation when editing.
On large collections, loading graph is very resources consuming. It would be great to not load all the data at every configuration change.
A button to toggle this behaviour would be great.

In filters for dates if we apply relative dates it shows from Sunday-Saturday, but we want it to show Monday-Sunday,

For our specific chart requirements (comparing different entities), it is imperative that we are able to plot the multivariate data in a spider chart.

I have read on the developer community that "...we are also planning a spider/radar chart in upcoming months." (dated 1 May 20) but since then I have been able to find any other reference wrt the realization for this feature.

The filters applied at the dashboard level should also be applied to the "Get Charts Data" export feature. The export currently contains all data from the underlying dataset.
This counters the user's expectation, which may be to export only what is filtered/visible.

When we embed a chart using the SDK, if I am using a Line chart which has a date on the axis, it is formatted in English locale only.

I would like to be able to toggle that between English and French as our project has a language toggle.

I want to keep hand on my data and prefer to run mongodb database on my servers.

Mongo charts is not so easy to maintain compared to mongoDB so i am interested in the cloud solution, but connected on my self hosted databases.

Oftentimes there are outliers in the data I want do visualize (e.g. all data between value 1 and 10 but few very large values like 100). This makes e.g. bar charts unreadable when binning because I have to decide between large bin size (hiding details in the parts of the data which are most interesting) or having the bars squeezed until they are not readable anymore. It'd be perfect to decide an lower/upper limit when binning and to summarize values below/above as "others".

I want expose some month numbers of my company, but the button "Force refresh" broken the numbers.

If it were possible to allocate teams to dashboards and data sources then management of a group of users who should only have access to a subset of charts would be much easier.
At the moment when a new user joins a team they must be added individually to each dashboard and datasource.

When I view the max/min aggregate in the charts, I would like to see the corresponding document's field. For example, I want to visualise a table like below,

_id              max(field1)               field2_of_maximum_field1

When we have an array of type [String, number], the second item is considered as string and hence, not able to do aggregation like sum.

For example, consider
```

$project: {

  transposed: {

    $zip: {

      inputs: [["aaa","bbb"], [1, 2]]

    }

  }

}

I am not able to create a data table with transformed.0 in col1 and max(transformed.1) in col2

When looking at server charts in Ops Manager METRICS, it would be good to have the ability to drill down to the database level. For example, I can easily see #connections to a host via Ops Manager UI. But, I want to know which databases in the cluster are being connected to, and currently, I have to mine the mongod logs for that.
This is discussed https://support.mongodb.com/case/00687156

The auto-placement behaviour of charts on dashboards may have its uses, but often it can be highly disruptive and require all sorts of chart-sizing workarounds and wasted time repositioning charts after they exercise their current anti-gravity tendency.

Can we have a fixed-placement mode so that charts stay where we put them on dashboards?

Add the Chart Title to its Loading dialog.

Would solve some UX problems for embedded charts - letting the user know what is coming to fill the space. There are workarounds for this currently (i.e. not using the charts title and having it on the render end instead... or having a conditional display that waits for the chart to load) but it would seem a more integrated experience all around to add context relevance to the Loading dialog, if possible.

Thanks!

It would nice if there was an option to toggle row numbers for text charts. This would be helpful, for example, when the rank of entries is important.

My suggestion is to allow for community-created visualization plugins, also called custom visualizations.
This type of functionality is offered on popular visualization tools like Grafana, MS Power BI and Google Data Studio.
In this way, you can leverage the power of the community to create new visuals for your Charts tool.