Correlation is a measure of a kind of relationship between two variables. It helps answer the question:
“When a variable A changes, does variable B tend to change as well, and in what way?”
A positive correlation means that if A increases, then also B increases; a negative one means that as A increases, then B decreases.
Even if correlation doesn’t mean causation, knowing when two variable are correlated between each other is important for prediction.
Let’s say the age of a driver and the number of claims are negatively correlated, meaning that the younger the driver, the more claims are filed, this is useful information to charge an higher premium to younger drivers, even if it doesn’t mean that the age is the cause of the claims.