Usage statistics - differentiating between the test plans and real plans

[magdalenadrafiova]

Currently the usage statistics dashboard does not differentiate between the test plans and the real plans. Could we differentiate these for the administrators?

Let me know if you need more detail about this.

[briri]

Yes. It is easy for us to determine which of the plans are 'test'. Do we prefer to exclude them entirely from the statistics calculations or just indicate how many of the total number of plans are 'test'?

[magdalenadrafiova]

If we could specify which are the test and which are total numbers that would be fab 🙂

[sjDCC]

We don't need lots of detail on test plans so exclude from total statistics. If we include anything on test plans just a total number is sufficient

Check whether we need to run a new rake task for old stats or see if it does this automatically

[sjDCC]

Make it a toggle so people can see both stats

[mariapraetzellis]

Add a checkbox which is defaulted to ticked to include test plans in stats.
Allow org admins to untick the box
Add text to show total plan figure to show this includes test as shown in screenshot below:

[sjDCC]

Suggestion from @xsrust that we may need two stats tables - one with test and one without

May be best to word it as 'exclude test plans' from count so overall

Alternatively say total number of test in parenthesis so we don't have to build in new functionality and logic

[xsrust]

If it's possible to have a number of test plans in each statistic so we can re-compute the totals with/without tests rather than double the entries in the stats table

[sjDCC]

Have validated with @mariapraetzellis and this approach is fine

[sjDCC]

Couldn't propogate test plan filters through to click-through pagination of results per month

Numbers generating for stats weren't discounting inactive plans but Rails counts are. This leads to a discrepancy between historical data and what may be showing now e.g. is a user has deleted plan since

[sjDCC]

Looks good. Easy to use filter and don't think anyone would notice discrepancy between numbers in graph and in click through. We already alert this is historical data

Closing out. Thanks @xsrust