If your peaks are well resolved, use the TIC. If you collect signal for the lower masses in the spectrum, this will do a good job of giving you areas that are roughly proportional to the concentrations of compounds present.
Beyond that I'd have to know which GC-TOF data analysis program you are using. There is the software that comes with the instrument and some people are using third party software.
If you are using a LECO Pegasus, I am willing to look at a data file or two to see if I can help you out. (I have to do this at home rather than at work - so I can not look at many files.) My e-mail is shown in my profile and at the bottom of this posting.