I find the Streaming export documentation very ambiguous, it doesn't go fully into detail about what data I will be working with once I start this type of export, so I don't know what to expect.
When the daily type of export is not an option because of data size, whatt remains is streaming the data into BigQuery, but streaming, besides additional cost, has some major limitations - traffic name, source and medium data points aren't included with this type of export, and these are the crucial data points in Google Analytics.
What I find confusing is this quote from the page linked above:
User-attribution data for existing users is included but that data requires ~24 hours to fully process, so we recommend not relying on that data from the streaming export and instead getting user-attribution data from the full daily export.
Can anyone who has tried the streaming export confirm if this means that user-atribution data will not be available in the "intraday" tables, but will be available in the "daily" table? If so, does this mean that the daily table can gather more than 1000000 events per day in this case? And do we need to have both streaming and daily export turned on to be able to gather this information?
First of all, you can find a sample dataset of GA4 export here.
For most of the use cases Daily export is enough but there is a delay between data collection and exposing data in BQ. If you need data even for the current day you will find it in the intraday table. Daily table can export more than 1M events per day but you will need GA4 360 (paid version). Streaming export (intraday) does not have this limitation.
Usually you do not need the user traffic source/medium. These user dimensions are just the first visit traffic source/medium. I assume you need session level traffic source/medium and it is still available in the intraday (streaming) export. Eventually you can calculate the user first traffic source/medium from there.