If the S3 bucket does not exist when the pipe is run, but the identity the pipe is operating under has rights to create a bucket, the pipe should create the bucket assuming there are no name conflicts.
This would increase the ease of use of the pipeline and make it easier to implement.