The main task for almost any company is making profit. It is rather clear. However, to understand at what price you should sell your product or how much can you spend money to attract users, you need to model your economy. Modeling will help you to understand what metrics (indicators) you should improve to get the maximum profit.

A beautiful sunny morning. We are happy as never were. In the last days we saw how the average check was growing. Yesterday we had 10 sales for $5, 5 sales for $10 and 4 sales for 100$. Mean was $26.3. It would seem that there were no signs of troubles, but in the evening the situation has changed dramatically. Mean was only $17.6. Nine dollars difference? Why? Because today we have only 2 sales for $100. In this case (to determine the real 'mean') we should use median.

We can't go past CASE WHEN clause especially when we are speaking of the data aggregation. Despite the fact that the SQL itself is not a programming language, it gives us a great opportunity to use control flow in queries without recourse to the stored procedure. Disadvantages of stored procedure are quite high threshold of entry and inability of writing the cross-platform code.

During data analysis, you can often face with several problems. You need to choose the right hypothesis test or to select the correct descriptive statistics. Here is a set of roadmaps and schemes. Feel free to use.

Hello everyone! Some time ago me and my partner started a new project which should conquer the world. Despite the fact that we are fully confident in its success, investors did not feel the same. And we had to make a business plan. It should reflect our vector of development that we want to get in the future and how we will achieve it.