What kinds of things have an influence on the size of a map package (*.mpk) that is created? I have noticed that when replacing a layer whose data source is an *.sdc file with the same data in a *.shp file it causes the size of the package to go up dramatically, so I'm wondering if there is any documentation on what contributes to MPK size? In all cases I do not check the "reference all data" box when creating my MPK, so that at least remains constant.
It might be a good idea to move this thread to a place like https://community.esri.com/community/gis/managing-data?sr=search&searchId=7db7b8e2-d378-4bfc-a9bf-30... or just https://community.esri.com/community/gis?sr=search&searchId=4ea0ac7c-a636-4734-a097-564fc63113c8&sea....
Shapefiles are uncompressed and can take up lots of room. Using things like a file geodatabase, or any sort of geodatabase can dramatically help reduce the size of a map package.
So that was one of the first things I tried. I copied all of my feature classes referenced in the map into a .gdb (I used ArcCatalog for this so it wouldn't save the geoprocessing outputs to the map when I packaged it since I also read that this could affect size) and then changed the source of the layers to use those instead. It only reduced the size of my map package by about 15 KB, so essentially not at all.
I had no idea where to ask this question so I will try to move it to Managing Data as suggested.
There are some tips and such in these documents that might help:
Scroll to the "Preparing Your Data for Sharing" section of this article.
You can use the
georeference tool for a finer degree of control
This resource has a python script that you can run that might help as well.
do you have images in tiff or jpeg format? other image formats take up less space (ie *.png)
If you have featureclasses or shapefiles that are just there to look nice and the attributes aren't important, just delete any fields that you don't need.
Pretty hard to discern what is taking up the most space, but a quick look with Windows Explorer in the folder where you have stored your data would give you a good idea which files and/or folders are the biggest... target those first by either removing unneeded fluff, reducing extents of input files and getting rid of unnecessary attributes are some of the obvious ones. Also check your geodatabases and/or folders, often they have content that you can no longer remember what they are or why they are there... perhaps intermediate processing results? A good house cleaning doesn't take long.