We are developing an empirical spectrum synthesis technique that yields independent measurements of starspot filling factor, f5, and starspot temperature, TS, by fitting TiO bands of differing temperature sensitivity. The absolute depth of the TiO bands constrains fS, while the ratio of their depths is a function only of TS. One strength of this technique is its ability to determine the spot parameters in traditionally difficult cases: slowly rotating stars, uniformly spotted stars, and stars that always have spots. For this initial study, we have used a simpler procedure of measuring the band depths in the most spotted star in our survey (the single-lined RS CVn binary system II Pegasi) and for a full grid of comparison stars (inactive G, K, and M dwarfs and giants). This yields TS and fS for a given assumed temperature of the nonspotted photosphere, TQ. The latter was further constrained by the use of simultaneous photometry. We have analyzed a series of spectra of II Peg obtained throughout a single 6.7 day rotational cycle in 1989 October. We find that starspots on II Peg are better modeled by comparison spectra of giants than by dwarfs. Combining TiO analysis with contemporaneous photometry, we find that cool starspots (TS ≍ 3500 K) are always visible, with a fractional projected coverage of the visible hemisphere varying from 54% to 64% as the star rotates. The nonspotted photosphere has a temperature TQ ≍ 4800 K. Our results imply that even at the historical light maximum of V = 7.2, at least 34% of II Peg was covered by starspots.