ATLAS.ti is currently one of the major qualitative data analysis programs, along with Dedoose, MAXQDA, and NVivo. They all have a learning curve involved, but they each also provide high-quality video tutorials to help you with that.
If you are considering doing your analysis by hand, the main thing I would think about is how large your dataset is. In particular, if you have fewer than 10 interviews, you would probably be better doing the analysis manually, and if you have 15 or more I would recommend using software. In between, it is up to you.
In short - Atlas.ti is worth learning as a QDA program; about on par for cost; good transferability between platforms (Windows:Mac:Other); straightforward to use; more powerful and more detailed than many programs.
Slightly longer - I found Atlas.ti to be relatively straight forward, but extremely powerful when you take its documentation on and practice with it deliberately. Your mileage will vary. Its capacity to be cloud-based and accessible is about par for the course for conventional QDA programs. So, if you plan to use Atlas.ti with a team, it is necessary that you read its documentation to reduce catastrophic overwrite errors.
It depends on language of transcripts, text, etx I would say atla.ti and MAXQDA are suitable for all languages compared to NVivo! However, for images and visualization and content analysis in languages using Latin alphabets. Atlas.ti has an old GUI and you open lots of menu and sub-menues! It doesn't support some file formats!