A quick answer is that living in groups incurs both individual and social costs, and only if the benefits of living together, whether evolved through kin selection (benefit: shared genes and thus indirect fitness) or through reciprocity (mutual cooperation), outweigh the costs will eusociality evolve. The other reason is that it is not quite easy to fulfill the three requirements for eusociality, namely overlapping generations, reproductive division of labor, and cooperative care of the young.

Sir, thank you for answering my query. Finally, someone noticed it. Another follow-up question: what made the termites, coral-dweller shrimps and other eusocial taxa have the criteria for eusociality to occur in their colony. Or what made them realize that they have to take care of their young, (is this simply kin selection), have a single reproductive female, and have an existing overlapping generations. Is acquiring all these characters a matter of chance that was strengthened by evolution as time goes by?

I believe that the acquisition of the "criteria" that lead to the characteristics that pertain to eusociality (one of which is protection of young due to overlapping generations) is partly influenced by the current traits of the species in question (for example, coral reef shrimp live in great numbers in a sponge and as such and they also possess a mechanism for defense/attack which is their major chela—thus, these species characteristics line up to result in eusociality in terms of protection of young). Of course, if these shrimp somehow evolved or came to not possess this chela, the older shrimp would not be able to protect the juveniles from invading species, and there would most likely be no eusociality present.