Train Your Workforce to Thrive in a Data-Driven Age

WebinarNovember 2021

The future of work is data. As the amount of data in the world grows at an exponential pace, learning data science skills is increasingly important for professionals in all industries to make data-driven business decisions. To meet this need, learning professionals need to rapidly develop scalable, high-quality data science training programs and use data to measure the effectiveness of their training programs.

This webinar begins with a discussion about how data is changing the world. Adel Nehme from DataCamp will provide examples of how employees of all types are learning and utilizing data science skills to positively impact their businesses.

Next, Sheil Naik from Bloomberg will talk about how his team partnered with DataCamp to deliver a blended learning program focused on teaching data technologies to hundreds of employees in Bloomberg’s Global Data organization from around the world. Finally, Sheil will show how his team used the same analysis techniques taught in Bloomberg’s Data Analysis with Python program to analyze Python usage data to measure employees’ behavioral change at Bloomberg.

Summary

As organizations grow more reliant on data for decision-making and strategy, employee training in data science becomes increasingly significant. This session underscored the necessity of making data science education universally accessible within organizations. It highlighted data science as a method for addressing business problems through descriptive, predictive, and prescriptive analytics. The webinar offered insights into Bloomberg's use of data science in business decisions and process enhancement, showcasing the effectiveness of mixed learning programs that merge online platforms like DataCamp with in-person classes to strengthen employee skills in data analysis using Python. The session stressed the need to link training to business goals and to use data in assessing the success of training programs. It also explored Bloomberg's innovative applications of data science in fields such as natural language processing and automated news production, emphasizing that data literacy is essential for everyone in an organization, not only data scientists.

Key Takeaways:

Data science is a fact-based method crucial for solving business problems.
Bloomberg employs data science to enhance business decisions and processes.
Mixed learning programs effectively strengthen employee skills in data analysis.
Data literacy is indispensable for all roles within an organization, not only data scientists.
Training should be linked to business goals to maximize effect.

Deep Dives

Data Science Methods

Data science is a fact-based method that utilizes computational and analytical tools to tackle business ...
Read More

challenges. It is divided into three types of analytics: descriptive, predictive, and prescriptive. Descriptive analytics aim to make data seen and accessible via dashboards. Predictive analytics use algorithms and machine learning models to anticipate future trends, while prescriptive analytics assist in making informed decisions based on data insights. This framework is supported by a data science workflow that involves accessing, cleaning, and analyzing data to produce insights. The abilities required for this workflow range from basic data literacy to advanced programming, making it crucial for everyone in an organization to be data-literate. As Adele Neme stated, "Data science is ultimately a fact-based method for solving problems using computational analytical tools."

Bloomberg's Data Science Initiatives

Bloomberg uses data science in innovative ways to enhance business outcomes and improve processes. Natural language processing (NLP) is employed to improve client searches on the Bloomberg terminal, ensuring that users receive the most relevant news stories swiftly. Additionally, Bloomberg's open-source contributions, such as the Python plotting library bqplot, enable analysts to create interactive visualizations. The company is also pioneering automated news production, generating data-driven articles without human intervention. As Shell Naik, Bloomberg's technical trainer, explained, "We're seeing people use our data science and data analysis skills outside of just those data science groups, and then using it every single day to make an impact at Bloomberg."

Mixed Learning in Data Training

Bloomberg's mixed learning program for data analysis with Python merges online courses from DataCamp with in-person sessions led by technical experts. This approach offers a consistent learning experience across global offices, allowing employees the flexibility to learn at their own pace. The program's structure includes basic online learning, in-person classroom sessions, and a final project using Bloomberg data, ensuring practical application of skills. Since its launch, over 450 employees have completed the course, many starting as complete beginners. This method not only scales effectively but also personalizes learning, enabling employees to apply data skills directly to their roles. Naik noted, "Mixed learning classes have a more consistent learning experience than a class that's just offered in person by an instructor."

Assessing Training Success

Bloomberg uses a data-driven approach to assess the success of its training programs, employing the Kirkpatrick model of evaluation. This model assesses reaction, learning, behavioral change, and results (ROI) to determine the effectiveness of the training. By analyzing data from their BQuant platform, Bloomberg measures changes in "producer activities" to gauge behavioral change, confirming increased usage of data analysis skills post-training. This data-driven evaluation ensures that training is linked to business objectives and shows tangible improvements in business processes and decision-making.