Can't create custom variance metric[ Edited ]
July 2017 - last edited July 2017
I have a time series graph containing the maximum 10 series (goal completion locations) from a possible 200+, currently sorted by goal completions (descending).
I want to be able to choose the 10 "most interesting" series by applying a filter with a custom metric. For example, I would like to be able to filter out any goal completion locations with a variance in daily completions below a certain level over the last 30 days until only 10 remain.
I have tried to create a metric which is the variance of the 30 data points, but Data Studio treats Goal XX Completions as an aggregated metric and I can find no way to treat each day differently as a result.
As a last resort, I have also tried to create a metric which is the range of daily Goal XX Completions over the last 30 days so that I can choose the 10 events with the biggest range, but I had no luck here either.
Apologies if any of this is unclear. I would be immensely grateful if anyone was able to help me sort this out so that I can get on top of optimising my conversions.