Happiness and well-being
In utilitarianism, the assumed goal of ethics and morality is the greatest amount of happiness and well-being for the greatest number of people.
How do we know that this is the goal? Evolutionary psychology.
Our brains are a neural network based machine learning device designed for survival and reproduction. In machine learning, there are two mechanisms for producing a learning effect. Reinforcement, which effectively strengthens connections between neurons, and punishment which effectively weakens them.
Our brain's reward system causes actions that result in pro-survival outcomes to be reinforced, strengthening the connections between neurons that fired to create this action and increasing the likelihood that the action will happen again. Those that result in anti-survival outcomes are weakened so the negative action will be less likely. Humans experience these effects subjectively as pleasure and pain.
It follows logically that our brains are wired to maximize behaviors that result in pleasure and minimize painful ones.
Though the evolutionary goal of this wiring is survival and reproduction, the brain itself only responds to the internal reflexive pleasure and pain reactions regardless of their actual survival benefit. Hence our goal is not to maximize our lifespans and population, it is to maximize our experience of pleasure and pain.
When we live a life that gives us robust and varied sources of pleasure, while minimizing pain and hardship, we experience this as a general sense of well-being and happiness. Therefore, the utilitarian goal of maximizing happiness and well-being for the greatest number of people can be derived logically from the observation of our brain's neural network.
For those that prefer a metaphor from pop culture, it's basically the point system from The Good Place.