Theory has predicted, and many experimental studies have confirmed, that resident plant species richness is inversely related to invisibility. Likewise, potential invaders that are functionally similar to resident plant species are less likely to invade than are those from different functional groups. Neither of these ideas has been tested in the context of an operational prairie restoration. Here, we tested the hypotheses that within tallgrass prairie restorations (1) as seed mix species richness increased, cover of the invasive perennial forb, Canada thistle (Cirsium arvense) would decline; and (2) guilds (both planted and arising from the seedbank) most similar to Canada thistle would have a larger negative effect on it than less similar guilds. Each hypothesis was tested on six former agricultural fields restored to tallgrass prairie in 2005; all were within the tallgrass prairie biome in Minnesota, USA. A mixed-model with repeated measures (years) in a randomized block (fields) design indicated that seed mix richness had no effect on cover of Canada thistle. Structural equation models assessing effects of cover of each planted and non-planted guild on cover of Canada thistle in 2006, 2007, and 2010 revealed that planted Asteraceae never had a negative effect on Canada thistle. In contrast, planted cool-season grasses and non-Asteraceae forbs, and many non-planted guilds had negative effects on Canada thistle cover. We conclude that early, robust establishment of native species, regardless of guild, is of greater importance in resistance to Canada thistle than is similarity of guilds in new prairie restorations.