Hey @enwanne 
That’s an interesting use case!

As you might already know, you can change the timezone of your project to whatever you prefer for UI and charting purposes, but the inherent raw data ingested in Amplitude is always in UTC.

From what I understand of your end goal, a couple of things I can point out -

• Afaik, there isn’t a direct way of comparing equality of a property value to another property using the dropdown. The only timestamp property you can select currently in the property dropdown is the “Server Upload Time” , which is the UTC time the event was ingested by Amplitude.

• 

  This is mostly accurate and on par with the client_event_time raw datapoint up to a few microseconds I believe, but can vary.  You can try different combinations of manually comparing a timestamp value to your scheduled_meeting_time and the server upload time if that helps to some extent. You always have to option to change and select the time interval in the metrics module to your choice. 

Coming back to your goal of “we want to know if users are performing a specific event during their scheduled meeting time”, the first thought which comes in my mind is doing a funnel analysis to answer this in a slightly different way. 

You can try building the funnel using the “Join/Start a meeting” event as the first event, followed by “Any Active Event” or whichever event you want them to perform. Since you want to know specifically for their scheduled meeting time, you will have to hold constant the meeting_id property. Let me know if this approach helps you answer your question to some extent.

This is very helpful. i didn’t think to use the funnel but that helps me get closer to my goal. Thanks!

Just to re-iterate how I understood this: you want to be able to analyse whether users start their meetings as planned or not.

The easiest solution would be to define your data so that you are able to answer the questions you want to answer. In your case, I would add new event properties to your join_meeting event (or equivalent): matching_meeting_id: yes/no, and time_difference_to_scheduled_meeting_time: (integer in minutes). This allows you to focus your analysis on these properties. And if you want to see how many people miss their meetings completely, you can either compare the event counts or create a separate event meeting_not_started_during_planned_timeframe that is triggered when the scheduled meeting time ends.

((Edit the names to match your schema.))

When you run into things that are challenging to do in Amplitude because of the limitations of the web UI, it’s often practical to do the hard work already in event generation.

This is where I landed after speaking to some of my team’s engineering counterparts. Glad to have the validation that I was headed down the right path. Thanks!

Another simpler method could be charting a time-series graph of your event meeting_id grouped by the property scheduled_meeting_time. Build this on an hourly graph - and you’ll see the which scheduled meet time is being triggered when. If the timings match - then that’s correct. If they are off then it’s not being followed.