People charged with maintaining the populations of endangered species are always worrying about how many animals are needed to make up a viable population. This also applies in nature: a species with only a few individuals is on its way out (the thylacine may be in this state today, if it's not already gone).
A common postulate is the 50/500 rule. The idea is that mammals need 50 genetically contributing individuals (that is, population N minus the nonbreeding animals (too old, too young, etc.) has to = 50 or more) to be viable in the short term, and over 500 to ensure long-term stability in the face of random events like epidemics.
Fewer than 50 animals does not automatically equal death. Northern elephant seal numbers may have dropped to around 20 at one point due to overhunting. Cheetahs passed through a natural genetic bottleneck that may have cut the population down to something similar. The European bison or wisent was reconstituted from six animals, although in this case the population was closely managed by humans to reduce the level of inbreeding.
A 2009 study by University of Adelaide and Macquarie University researchers suggested that neither 50 nor 500 is adequate. They came up with the figure of 2,000 animals in the face of large-scale stresses like climate change and loss of habitat. These factors vary with species and location, but the study is still concerning to wildlife managers. A 2008 study from the University of Boulder had already suggested that models tweaked for individual species, widely used in conservation programs, may be off by a factor of 100. See:
What does that mean for the animals of cryptozoology? If the existence of an animal like the sasquatch is uncertain, its population is more so. If there are only a few dozen sasquatch huddled in the Pacific Northwest, their chances for survival are low but not zero. Primates, which have few offspring and take care of them closely, may be able to adapt to that for a long time. A population in the thousands is much safer, although the likelihood of the existence of an animal for which we have no physical remains would seem to fall if one is postulating there must be thousands of them. It's all more complicated than I am qualified to sort out beyond these general thoughts, as you then get into how many large primates the presumed habitat could support, odds of accidental discovery (hunting accidents, roadkill, etc.), and so on. It varies with every presumed species: for example, the yeti's habitat, the high forested valleys of the Himalayas, is much smaller than potential sasquatch habitat but also sees fewer human residents and travelers.
The point here is that figuring out how many animals a species must include to survive is not an exact science, whether you're dealing with a known or unknown species. The implication is that this whole conservation business is trickier than we thought, and conservation of habitat, the major safety factor, rises even higher in importance. Let's hope we have the wisdom and resources to make it all work.