The aim of this project was to analyze a dataset regarding ride sharing data. This data covered from city type, to date, time and fare for each ride.. The main tool used to develop this project was python 3 through Jupyter Notebooks. Now, let's dig into it...
The analysis must start with a break up of the percent of rides per city type. Notice how urban cities have a greater percentage of rides; this may be due to the number of people in the location or availability of drivers as well.
Now, the data answers this same question. Notice, in the following graph, how there are significantly more drivers in urban areas than in suburban or rural areas. This can be one of the main reasons for having more rides in urban areas.
However, given the fact that there are fewer rides in rural and suburban area, the price per ride increases significantly. This is an inverse relation, meaning, the more the number of rides, the less it costs.
Despite the fact that trips in rural cities cost more, the total fare is still significantly higher in urban cities. This is due to the high number of rides in urban areas. 65% of all the rides stored (2375) comes from urban cities.
From our data we are able to tell how rides behave in each city type. Through this we can have focus groups in order to maximize revenue in each type of city, separately. I would give 3 recomendations to the CEO for addressing any disparities among the city types.
- First, focus on where the higher concentration of rides is (Urban cities). Focusing on each city group separately can allow the company to have a better approach on marketing campaigns that will increase revenue.
- Try an A-B testing experiment. In some suburban cities, make more drivers available, and in some others keep it the same way. With this experiment, the CEO can analyze how is the ride behavior in suburban cities (our middle group) and check for a revenue increase.
- Charge more per mile in urban cities given the fact that rides in urban cities are significantly shorter.