final_project_part_1

Final Project, Part 1: Pitch
Test and validate 2–3 initial ideas.
In the field of data science, projects are practical. One of the best ways to quickly test ideas is to share your ideas with others and 
get feedback. Start by reviewing the requirements and suggested data sets, then generate a list of 2–3 potential ideas you'd like
to pursue.

Requirements
Generate problem statement (of at least 250 words).
State the questions you seek to answer, data set suggestions, and possible challenges.
Create a written pitch that can be shared with your instructor and classmates.

I plan to use order data from an e-commerce company around interaction volume: how many phone calls, livechats, emails, and SMS
are handled on a daily basis.  I will have access to 2-4 years of data (with timestamps down to the second), depending on the 
interaction channel.  I would like to look at this information, which will have dates, counts, times, interaction channel, and maybe
some other information like category and reason tags, for further breakdowns.  I would like to see if there are trends and predictions
to be made around holidays / months. Also, if possible, I will compare this to order volume and see if we can do demand forecasting 
using last month’s order volume and interaction counts to determine staffing for the following month.  

Possible Areas to Examine:

How does interaction volume change with seasonality, month to month, year to year?
How does time of day affect interaction volume?
How does time of day affect order volume?
How does interaction volume vary by day of the week?
Growth rate of interaction channels
Correlation between category/reason tags and seasonality
Correlation between order volume and interaction volume - is order volume a good predictor for interaction volume?
How can we use this information for demand forecasting / better staffing optimization?