Analyzed the Starbucks dataset in order to help Starbucks make better decisions to increase its revenue and customers.

Following problems were identified in the sample dataset based on the research.

-High prices with no customer rewards.
-Multiple branches within small geographical radius.
-The use of semi-automated machines instead of fully automated machines eventually lost the brand's authenticity.
-Lack of hospitality training to employees.

We can conclude that there are about $7.5 of mean purchase amount difference between 'non_offer' and 'offer'. Although engaging the promotions needs the minimum required purchase amount(a.k.a. 'difficulty' in our dataset), the difference is pretty significant. 

From our Predictive Analysis, we can conclude that as there are many men who crave for discounts, if we increase the offers and increase the minimum required purchase amount slightly it can result in generating more revenue and profits.