Embedding Teams into a website

maynardsAH 

The query will be many magnitudes faster if you use the Usage table (which has already aggregated the usage data), rather than trawling through a massive number of records.  

If you are not worried about whole month, this is a simple query for each 30d period

Usage
| where TimeGenerated between ( startofmonth(now(),-3).. endofmonth(now(),-1) ) 
| summarize GBday = sum(Quantity)/1000 by bin(TimeGenerated, 30d)

Go to Log Analytics and run query

 or to have whole months, maybe this is a start:

union
(
    Usage
    | where TimeGenerated between ( startofmonth(now(),-3).. endofmonth(now(),-3) ) 
    | summarize GBmonth=sum(Quantity)/1000 , min(TimeGenerated), max(TimeGenerated), month = datetime_part("month", min(TimeGenerated))
),
(
    Usage
    | where TimeGenerated between ( startofmonth(now(),-2).. endofmonth(now(),-2) ) 
    | summarize GBmonth=sum(Quantity)/1000 , min(TimeGenerated), max(TimeGenerated), month = datetime_part("month", min(TimeGenerated))
),
(
    Usage
    | where TimeGenerated between ( startofmonth(now(),-1).. endofmonth(now(),-1) ) 
    | summarize GBmonth=sum(Quantity)/1000 , min(TimeGenerated), max(TimeGenerated), month = datetime_part("month", min(TimeGenerated))
)