What do you like best?
It is an advanced statistical tool. It has the power to run advanced algorithms on huge dataset easily. Also, It integrates well with other products like SQL server, Teradata and processes can be automated.
What do you dislike?
It is quiet difficult for beginners but once you get used to it, its power can be realized. It does not have good graphic capabilities. The packages which they have are expensive and not really fascinating to modify. Also, since it is not open source, there are lot of advanced analytical techniques it is not up-to-date with, which are out there and shared by industry experts.
Recommendations to others considering the product
It depends on how much knowledge my colleague has on data analytics and how much time can be devoted to learning a new software. If he/she is relatively new to data analysis and cannot spend a long time learning SAS coding, I would recommend him/her to use SPSS which has a very simple point and click interface. If he/she has data warehousing/SQL experience before and wants to dig deep into data, he/she probably is a good starter to learn data mining through SAS.
What business problems are you solving with the product? What benefits have you realized?
I have use SAS to perform advanced machine learning techniques like logistic regression, random forest and gradient boosting.
We predicted the number of people who are likely to churn in the next couple of months, so that they can be retained by giving special incentives thereby minimize losses.
The results obtained are robust. It can be used for predictive analytics, data exploration, Business Intelligence and Statistical analysis.