In this article, I am going to explain how bridging the gap between data scientists and engineers may help your company unlock the full potential of the data. I will demonstrate collaborative strategies that focus on sustainability and emphasise the relationship
between data science and sustainability based on the project titled “Step”. Step is a Sustainable Choice Program we implemented in Yapı Kredi Bankası (YKB), which is one of the largest commercial banks in Turkey. In 2023, it received a prestigious PRİDA Communication
Award in the Sustainable Communication nomination.
In this image: Yapı Kredi Step Project mockup
Data Science and Data Engineering: What is the Difference?
To understand the gap between data science and engineering, we need to look into the core functions of the two branches. Even though both focus on data, data scientists and data engineers usually have different goals.
Data science focuses on extracting insights and patterns from data and is responsible for optimal and reliable data storage, data transformation, and fast and convenient access to data. Data scientists create and train predictive (and beyond) models using
machine learning algorithms and neural networks to help businesses find hidden patterns, predict developments and optimise key business processes.
Data engineering aims at creating data pipelines and infrastructure. Data engineers directly address business needs by testing hypotheses and building predictive models.
Even though both directions go closely together, misalignment between them is not uncommon. This may result in problems and inability to turn analytical data into actionable insights. Bridging this gap between data science and engineering is essential to
make the most out of innovative approaches to data, and it mostly lies in establishing healthy collaboration between the teams.
Tactics for Collaboration
Building multi-disciplinary teams and foster collaboration is a well-known tool that helps ensure that both technical and business requirements are met. Ww established cross-functional teams of data scientists, engineers, and domain experts to tackle the
complex data handling needs.
Encouraging well-organised documentation and knowledge sharing between data science and engineering teams helped the teams better understand each other's workflows and requirements and greatly improved empathy and collaboration.
During the project, we encouraged the use of shared tools and processes, such as version control systems and continuous integration/continuous deployment (CI/CD) pipelines. This helped us identify and fix possible issues early and improved consistency across
data projects.
Lastly, we adjusted our Agile processes to promote iterative development and frequent collaboration between data science and engineering teams, allowing for quick feedback loops and rapid iteration.
YKB Step Project: Case StudyThe aim of the project
The YKB Step Project has become one of the works that have made a difference in the recent period.
The aim was to enhance sustainability practices within the banking sector through data-driven initiatives. We wanted to motivate those who did not know how to take action against major issues such as the climate crisis and those who see themselves as weak
for change, to make sustainable choices.
The idea of the Step project lies in converting customers’ sustainable choices into points. The actions include opting not to print an ATM receipt, choosing eco-friendly transportation or public transport, donating to nonprofit organisations, and more. The
name of the project lies in one of the activities, issuing points for a certain number of steps taken during the day.
With Yapı Kredi Step changes consumption habits, lifestyle and daily choices for sustainability. We enable our customers to donate the points they accumulate to social responsibility projects, and we take action together to turn the world into a better
place in many areas from education to the environment.
What Was Done
The primary goal of the Step Project was to integrate sustainability principles into YKB's banking services and operations, aligning with global sustainability goals and societal expectations. This required gathering, processing and analysis of vast quantities
of user data.
We brought together multidisciplinary teams of data scientists, engineers, domain experts, and business stakeholders so that they could seamlessly collaborate on the project.
The Step Project began with data collection and preprocessing, involving the integration of various data sources such as transaction records, customer demographics, and sustainability indicators. We needed a strong collaboration between data science and
engineering to analyse customer behaviour, identify patterns related to sustainable spending habits, and develop a sustainable solution that would find a way into customers’ daily habits.
Data scientists conducted exploratory data analysis (EDA) to identify potential features and variables related to sustainable spending behaviour. They used machine learning algorithms and data mining to gather data and analysed large volumes of transaction
details to extract insights related to spending patterns, then developed and trained predictive models for customer behaviour.
Based on the goals of the project, data engineers built machine learning models for analysis and created scalable data pipelines to efficiently and truthfully process the collected data. Engineering teams employed agile methodologies and DevOps practices
for successful development, deployment, and iteration of data pipelines and models. Scalable data pipelines and implemented machine learning models helped process and quickly and efficiently analyse large volumes of transaction data.
Data scientists and engineers collaborated to analyse customer transaction data, identify patterns related to sustainable spending habits, and develop personalised recommendations for customers. This collaboration also enabled YKB to launch additional sustainability-focused
products and services, such as green investment portfolios and energy-efficient financing options, driving positive environmental and social impact.
Obviously, due to the nature of the financial field, we had to incorporate best practices for data governance, security, and compliance to our solution. This was a challenge that required deep involvement of legal, information security, and other teams into
the process. This underlines the importance of fostering the culture of continuous improvement and strong collaboration of teams on many levels in order to create a successful product.
The Impact
This collaborative effort led to further creation of more complex sustainable banking initiatives: green investment portfolios, carbon footprint tracking tools, and energy-efficient financing options. The project enhanced YKB's reputation as a leader in
sustainable banking practices, helped attract new customers and foster long-term relationships with existing ones.
As of February 2024, the number of members of the Sustainable Choice Program STEP has reached 570 thousand. With their sustainable choices, STEP members have saved 50 million pieces of paper and saved nearly 4500 trees so far.
The perception of the company as "supporting society in social responsibility" score increased by 9% and the "provides guidance on sustainability" score increased by 10% compared to the previous quarter. This turned Yapı Kredi into the brand with the highest
increase in these metrics compared to its main competitors.
Through the insights gathered from the Step Project, YKB was able to tailor its offerings to meet the needs of environmentally conscious customers and bring attention of the masses to the environmental issues. We have implemented a project that guides consumers
on sustainability through its application, rewards users with points and enables them to use the points to donate to NGOs.
The success of the project inspired YKB to extend its sustainability initiatives beyond banking. Now we collaborate with partners across various industries to promote sustainability and drive environmental change, and we are eager to share best practices
and lessons learned from the Step Project to encourage other organisations to leverage data science and engineering for social good.