Building a Data Pipeline & Visualisation Platform

Building a Data Pipeline & Visualisation Platform

Company Sponsor: The Lo & Behold Group

The Lo & Behold Group is a hospitality company that creates, owns and operates a series of timeless, thoughtful concepts, each with a unique story and a distinct perspective on the cultural-culinary landscape. Across all, the single thread that binds is a carefully considered sense of place, purpose, pioneering design, and a commitment to finding magical moments for everyone.

Title: Building a Data Pipeline & Visualisation Platform

Project Team: Alex Chong Ee-En, Bryce Jeremael Ong, Joline Aw Juan Lin, Rebecca Yang Ru Shi and Warren Chia Wei Wen

Problem

Each restaurant operates on an independent basis with different POS systems for business operations. Data collected from the different restaurants are not used effectively to draw insights about customer spending behaviors.  Existing dashboards are operations-based, disorganised and unable to provide comparison between different restaurants.

Solution

The way we approached this project was to completely understand the needs and wants of the client, as well as continuously iterate through the solution, ensuring that everything would go smoothly during the handover. This culminated in the main theme of our methodology being functionality and future ease of use.

The first thing that we set out to do was to completely understand the pain points and key requirements of the client, as well as the industry that they operate in. As such, we met up with the finance department multiple times, making sure that we talked to everyone on the team to understand their level of comfort with using the stack that we proposed, as well as their needs and wants with regard to this solution. After communicating and understanding their viewpoints and requirements, we were better able to craft a solution that relieved their pain points.

In building the final dashboards, we sought to understand how these dashboards would be used in order to ensure that they would not be rendered useless if any minor changes are made. Hence, our building process was an iterative one, where we built various prototypes, taking into account the client’s feedback as well as fixing any errors along the way.

Finally, we wanted to ensure that our solution would be useable by the clients. Given that they did not have any experience with R, we had to come up with a suitable handover document in order to ensure that they would be able to replicate what we had done over the time period in which our solution was to be used. Hence, we created a comprehensive guidebook which covered everything from software installation to data extraction. This would ensure the usability of our solution in the long run.

The showcase poster details the actionable insights for the client.

Click image to enlarge poster

Taking this capstone, first and foremost, as a team, we all learnt how to manage the client individually. This was not limited to the expectations and deliverables, but also how to communicate clearly and make sure that our points were received by the client. In addition to this, we picked up the technical side of things through self-directed learning of R Shiny dashboards and other data cleaning techniques in R.