Redundancy is where things are duplicated. Redundancy is often confused with backup. Backup is where you have a separate copy of your data, and there should be at least one copy stored in a different location.
Redundancy is where you have copies of things to ensure continuity. One place redundancy is often used is with equipment on a network. You may set up multiple paths through different equipment from one location to another location. That way if a device or line quits working you still have connectivity to the other location.
With data, cloud providers may often provide redundancy but all they are doing is storing the data in two different areas of the equipment at the same location. Then if a disk drive fails, the system will switch over to the other drive.
The problem with this data redundancy is that people think this means backup and if something occurs to the data or the data center then the data is lost. Although this has not been a problem with major cloud providers yet, if you only have your data redundant it could be lost. If it is not backed up and the data is corrupted (either errors enter, or data damaged or maybe ransom ware attack) then the data is gone. If the data center is destroyed (hurricane, tornado, fire, terror attack — one of things that some organizations lost in 9/11 in towers — etc.) then your data is gone if it is not backed up to another location.
Redundancy is important in keeping an organization going when equipment, lines connection to ISP, etc. fail, but does not cover when the center, etc. is destroyed.