Context : 

This project serves as another opportunity to practise and refine key data analysis skills, including data preparation, cleaning, analysis, and visualisation. Through the completion of this exercise, I tried to gain valuable hands-on experience in the field of data analysis. 

Project overview : 

The COOP in Berlin is a cooperative supermarket. Members can enjoy high-quality products at reasonable prices in exchange for 3 hours of their time each month. A dedicated software creates shifts where members can register to fulfill their duties.

Objective : 

A significant amount of data has been generated from this software, and executives at the COOP want to understand variations in attendance in order to optimize the allocation system.

Software used


Data sources : 

Link to data source

Tasks performed : 

Leading question : 

  1. How many shifts has a team leader ?
  2. Hypothesis: shifts that have a higher proportion of ABCD members have less absences than those with more flying members -> to be investigated
  3. Are there long term trends with regards to attendance rates over the last 2 years since data began? Does it vary by shift (morning vs. evening, weekday vs. weekend, is there seasonal variation in the data.
  4. What is the average shift staffing? 
  5. Which shifts are most popular? 
  6. Is there a pattern for individual members with attendance rates dropping? 
  7. Total work hours done by member and in total for specific time periods 
  8. Do some members do more than 1 shift per month? If yes, which ones? What is the distribution of member shift attendance by hour?
  • Cleaning data

Detecting and removing any duplicate entries, resolving miscategorizations, and addressing any blank or null data points. In this case it was very clean from the start.

  • Building the dashboard.

The visualisation stage of the project will prioritise readability and clarity, avoiding overwhelming the audience with too much information at once. 

To achieve this, the project will employ effective organisation techniques to present the data in a clear and digestible format, allowing for easy interpretation and analysis. 

Additionally, the visualisations will incorporate the ability to filter and drill down into the data, providing a deeper understanding of the underlying trends and patterns.


The analysis of this dataset has yielded several insights regarding shift attendance. 

Notably, the findings suggest there’s no significant difference in shift attendance related to allocated time (morning vs evening, week day vs week end.)

This study nevertheless demonstrated that there is considerable room for progress regarding allocation. Only 50% of shifts are allocated when there are a sufficient number of people to fill them.

Link to tableau