It is incremental aka saving the diff.
Another advantage of Duplicati (and all softwares implementing it) is deduplication: the same file present many times will only be saved once.
In my situation, a lot of users share files through various services, so deduplication saves a ton of space.
The downside is that if there’s corruption in the backup, all copies of the file are corrupted. But with two different backups + source files, I’m not too worried about this. You can also add parity files for error robustness to Duplicati I think.