I would like to add my thoughts on this as I have an opposite use case here.
There are some limitations put on Google sheets that say you can’t have more than 5 million cells. Also that your sheet will slow down a lot before you even reach half that amount. Thus when realistically looking at the amount of data gathered in a year’s use of appsheet when using multiple tabs in the same sheet, I felt it better to split that sheet up into individual ones instead.
Another reason for splitting them up is that I use the Appscript onChange() project triggers a lot to generate sequential order IDs for quotes, cases, references etc.
There are quotas placed on the amount of times a user can run a trigger in a single day and the onChange() event will fire on any change made to any cell in any tab in a sheet. So by splitting the sheets up into multiple files I can make my triggers only fire when my master table is changed.
onChange() is something I use massively for creating folder structures in Google drive and linking the URL back into appsheet as well, but I am looking forward to being able to trigger Appscript from within appsheet at some point.
I hope this alternative viewpoint on this question makes sense. I have a naming convention I use for my appsheet projects where if you were to view an app folder you would see a sheet called Appsheet - Project Name. Then a number of other sheets called child tables…Appsheet - Project Name - Child Table - Name of child table 1, Appsheet - Project Name - Child Table - Name of child table 2 etc. This also makes it easy to bring tables from other apps together and create apps that can share data.
Yes, initial sync times can hit 15 to 20 seconds, and you have to bear in mind I am using many virtual columns to perform calculations here, but I think it’s worthwhile for the time saving down the line with App sizes.