0
$\begingroup$

I work as a data analyst at a theme park (kind of like disney) but we operate only during 4-6 months in an year. I was asked to analyze the F&B sales at the different locations in the park and compare this year's sales with last year's sales.

Couple of constraints and issues:

  1. The number of outlets in an area keep changing every year. Last year we might have had 10 outlets in Area A but this year we only have 3. - To overcome this, I divided the sales by the number of outlets so I will be comparing the revenue per outlet for different years.

  2. Different outlets operate for different days. For example, this year, outlet A might have operated only for 30 days but outlet B would have operated for 90 days. So outlet B's revenue would be higher. - To overcome this, I divided the revenue per outlet by the number of days the park operated this year. So let's say the park operated for a total of 100 days then I will divide the revenue per outlet/100 to get the revenue per outlet per day. Someone from the management pointed out that it would be better to divide the revenue per outlet by the number of days each outlet is operating to get the exact figures. That approach makes sense but it seems a bit more time consuming. Would it be worth redoing the analysis to change this?

  3. Since we operate on a season basis, we operate for different days every season. For example, last time we operated from Oct'19 - Mar'20 but this time we operated from Oct'20-May'21. I thought that displaying the revenue per outlet per day would overcome this but someone from management mentioned that they would rather have me forecast the expected revenue from Apr'20 - May'20 so that we are comparing Oct'19-May'20 vs Oct'20-May'21. And then I can use this new forecasted revenue to get the revenue per outlet per day.

  4. Because of covid, we had lesser footfall and hence lesser revenue this time. Someone suggested to normalize this year's sales by taking the footfall into account - i.e. take the drop in footfall from last season to this season (let's say the drop is 30%) then take 70% of last year's sales to accommodate the drop in footfall. Is this the right approach to it? Wouldn't comparing the revenue per outlet per day take care of the drop in footfall?

  5. All of this needs to be in a dashboard and they want a functionality where they can make a like to like comparison - We will be reopening in Oct'21 and at that moment they want to compare their performance with previous season (i.e Oct '20). There are 2 problems with this:

    1. In Oct'20, we started on a monday (weekday) whereas in Oct'21 we might start on a saturday(weekend). Our weekend performance is higher than weekday so comparing monday vs saturday wouldn't be correct.

    2. As mentioned earlier we operate for different # of days across different season. For eg, this time we operated for 191 days whereas last time we operated for 161 days. Even if I compare the 1st sunday of last time with 1st sunday of this time and so on, I will not have anything to compare with after we finish 161 days this season. How do I accurately make this comparison?

Sorry for the long post but I have not come across such constraints before so I'm not sure how to deal with all of it. I would really appreciate any help! Thanks in advance!

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.