Explore Data Science
We paired space, data science, and instructional software in Booz Allen Hamilton’s first product.
Showing the foresight they built a reputation on as consultants, Booz Allen Hamilton (BAH) predicted that data scientists would replace the role of data analyst. Internally, they faced a training gap: New hires either had a pure statistics or a pure computer science background, and data scientists operate in both realms.
As fans of Code School, BAH sought Envy Labs’ help to deliver a technical education platform. Python sits atop the data analysis world, and a truly effective training environment would need to offer hands-on exercises (and feedback) in the language, the academics, and the delivery of their unique approaches and solutions.
- UX Design
- Narrative Theme
- UI Design
- Spinning Planets
- Ruby on Rails
- Python Execution
An interface geared towards clear, actionable feedback is key when training users new to statistics and computer science at the same time.
Explore Data Science marked BAH’s first venture into the world of product creation and software as a service. Keeping everything on track meant a reliance on our processes and experience.
By the client’s own admission, learning data science and Python could be a bit dry. Taking cues from our past gamification and storytelling experience would serve to liven things up.
Bonus: Let’s Make a Product!
Secret goal number four would surface midway through the build: With early-to-market content and technology, why keep this internal?
Weaving a narrative into the course content meant we were on the hook to pick a theme early. Inspiration in the form of Oregon Trail, sports, and microbiology were tossed around. In the end, a spaceship-building, space exploration theme won. Space always wins.
Interactive, Long-Form Content
Statistics and programming both favor the ability to reference examples and explanations freely. In support of this, we opted for a long-format experience: Combining the best parts of readable textbook-style content with interactive challenges placed exactly where needed.
When you combine challenging content with fun space-themed collectibles and friendly leaderboard competition, even the skeptics loosen up and have a little fun while learning.
The Planets Spin
Instead of presenting users with the usual array of chapter titles and lesson plans, we dropped them into an explorable solar system. Each level of Explore Data Science carried a unique theme: from Earth-like environments to inhospitable lava planets, each chapter is memorably distinguished from one another.
Prizes for Completionists
Trophies, leaderboards, and even a custom spaceship emblem designer — half-measures with themes fall flat, so we made sure the experience was consistent throughout the entire application.
Client-Friendly Content Creation
Content is the hardest problem in education, and we made simplifying the creation process a priority from the start. Text, media, and even challenges formatted in Markdown were automatically pulled into the platform.
As they progressed, users transitioned from the textbook-style content to a fully interactive Python executor as necessary.
Thanks to D3, progress visualization and generative motifs made the interface feel a bit more lively.
For new and old employees alike, Explore Data Science represented the first time BAH had assembled all of their core data science knowledge into one curriculum.
Similarly, the application stood as the company’s first foray into product — complete with a launch party at the New York edition of Strata Data Conference.
The product’s appeal and success led to a 2015 SXSW Interactive Innovation nomination.
Sale to Metis
Further validating the platform and curriculum, BAH sold Explore Data Science to Metis, a division of Kaplan, in 2016.
“Explore Data Science is … an exceptional tool for providing introductory training on data science fundamentals to a broad audience.”Jason Moss