The program uses the text on each layer in the motion file. All text layers will be listed as possible targets for the columns from the .csv file.

 

After the files are generated, you will need to render them using either compressor, motion or compressor server.