A single photo from a recent phone can be five megabytes or more. Understanding why makes it obvious what to change.
Resolution comes first. Phone cameras now capture twelve, forty eight or even two hundred megapixels. That is far more detail than any screen displays, and every one of those pixels has to be stored.
Format comes second. A photo saved as PNG can be many times larger than the same photo as JPG, because PNG stores everything losslessly. That is right for graphics and wasteful for photographs.
Content comes third, and it surprises people. A busy photograph of leaves or gravel compresses poorly because there is little repetition for the compression to exploit. A plain sky compresses to almost nothing. Two photos at identical settings can differ in size several times over for this reason alone.
Barely. Metadata is a tiny fraction of the file. The pixels are what take the space.
Lowering your capture resolution helps, but you cannot recover detail later. Shooting large and resizing copies is safer.
No.