How would you prioritize data points when resources are limited and you need to make a recommendation quickly?
