Posts

Showing posts from May, 2024

2024: Week 20 - SuperBytes Customer Data

Image
  Challenge by: Saampave Sanmuhanathan We're continuing with DS43's challenges so over to Saampave  t o explain the her next challenge.  _____________________________________ This week we are looking at the customers of SuperBytes. Specifically, Phill Down is interested in which customers spent the most with us on different days of the week last year.  Inputs The input data contains two sheets one with the transaction details of each customer and the other with customer names and customer ID number.  Customer Spending  Customer Names  Requirements Input the data Join sheets the remove names and maintain privacy Group Online and In Person into one field Add a field for the day of the week the transaction was completed Rank sales for each weekday Create a parameter to output a data set for each day of the week Output the data Add day of week to the title of the worksheet Output Monday example output: 7 fields Rank Customer ID Receipt Number Gender Online or In Person Sales Weekda

2024: Week 19 - Solution

Image
Solution by Tom Prowse and you can download the workflow here . Step 1 - Input Tables The first step is to input all of the tables from the input file. We want to create a union and stack them on top of each other so therefore we can utilise the 'Union multiple tables' functionality within the input step. From here we want to select all of the tables where the file name matches 'SuperBytes Sales_Profits.xlsx' and don't need to work about any wildcard values. As a result we should have the worksheets from various years input into the workflow:  Once the tables are in the workflow, we can rename some of the fields so they match the data that is in them:  F1 --> Quarter Table Names --> Year Then make sure the Year is a whole number. At this stage the table should look like this: Step 2 - Sales & Profit Values We want to ensure that the number values in the Sales & Profits field are formatted in the same way and take into account the string value.  To do t

2024: Week 19 - SuperBytes Sales and Profits

Image
 Challenge by: Saampave Sanmuhanathan We're continuing with DS43's challenges so over to Saampave t o explain the her first challenge.  _____________________________________ This week we are looking at the sales and profit data of Superbytes between 2018 and 2022. The finance department would like to compare the yearly sales and profit of the store. Inputs The input data consists of 5 sheets corresponding to each year (between 2018 and 2022). Each sheet includes sales and profit by quarter of each year. 2018 table There are 5 sheets in total with the title of the sheets according to the year the data was collected. Requirements Input the data Union tables together We want to remove the units from the sales field to convert the data type to the whole number Repeat this process for the profits field Aggregate sales and profits by year Output the data Output 3 fields Year Sales Profits 5 rows After you finish the challenge make sure to fill in the  participation tracker , then sh

2024: Week 18 - Solution

Image
Solution by Tom Prowse and you can download the workflow here .  Stage One Step 1: Animal & Plant Details First we need to input the Animal Details and Plant Details tables and combine them by using a Union step. We can then ensure that all the fields are correctly mapped by using the selections in the Union tool.  The miss-matched fields that need to be combined are Animal & Plant and Status & Stattus. These can be selected and then added together by using the + symbol.  We can then rename any fields and remove fields that we no longer required. The table should look like this:  Then using the Table Names field we can extract whether the row of data is related to an Animal or a Plant. In this case we can utilise the automatic split functionality where Tableau Prep identifies the part of the string that we want to extract and creates a new field for Animal or Plant.  We can then focus on the Class field to extract the English name instead of the Latin name. For this we can