That's a broad question. Who is the "we" you refer to? Is it the general public, reservoir operators, emergency or civil defence managers, farmers, construction project managers, etc. Are you talking about short lead time or long lead time forecasts?

Several examples:

The organization who manages the water supply for New York City (NYC) operates a very large system of reservoirs and interconnections. They have to ensure that the water the capture remains reasonably sediment and pollution free. They were faced with large capital expenditures to achieve new quality standards. They worked with the U.S. National Weather Service (NWS) and some private contractors to create decision support software that takes the flow forecasts and their uncertainties that allows them to optimize how they move water through their system, thereby reducing the cost of pouring concrete.

The system that NWS uses to provide probabilistic forecast information to the NYC decision support systems is a complex of atmospheric and hydrologic models. The various models can each be broken down into subsystems. For each forecast cycle, those subsystems are run multiple times with each run being initiated with states and parameters that vary according to appropriate statistical distributions. The resulting subsystem outputs are then "debiased" based on historical model runs. The information, along with uncertainties, is then passed to the next subsystem until the final output, which includes the accumulation of uncertainties. It's a very complex system and there's a lot of Bayesian math involved. The resulting probabilistic forecasts have lead times of days, months and seasons. They help water resource managers, farmers, etc, to better operate their systems

The weather forecasts NWS provides to the public, include a probability of rainfall (POP) along with amounts for several days into the future. The public can then make decisions like shall I paint my house or should I take an umbrella when I walk to work tomorrow. That's actually a tricky one because quite a few people misunderstand the statistical meaning of the POPs. But the public do pay a lot of attention to the POPs.

Hope you find this useful - Geoff