This post combines a series of LinkedIn articles written as part of the Data Next Level Challenge. You can find the original posts here:
- Part 1: Why I Started This Journey (Published: November 8, 2024)
- Part 2: AARRR Framework & Metrics Analysis (Published: November 13, 2024)
- Part 3: Implementation & Practical Insights (Published: November 20, 2024)
As part of the Data Next Level Challenge, I embarked on a three-week journey reading “Growth Hacking” to deepen my understanding of data analytics and connect with fellow data enthusiasts. This post compiles my complete review series, sharing key insights and personal reflections from this learning experience.
Why I Took This Challenge
Coming from a government background with no private sector experience, I’ve often felt uncertain about product sense and case study interviews. During my career transition, I’ve been contemplating my true career aspirations. While I initially focused on becoming a data scientist solving business problems, I’ve recently developed an interest in MLOps. I believed reading industry-focused books would help clarify my career path.
Key Insights from Chapters 3-4: The AARRR Framework
Core Principles for Metrics Analysis
-
Accurate Data Tracking
- Proper tracking of user acquisition sources (Google ads, affiliate marketing, referral programs)
- Emphasis on comprehensive data collection without omissions
-
Data Segmentation
- Breaking down data by cohorts for meaningful insights
- Importance of user grouping for revenue-based decisions
-
Customer-Centric Analysis
- Focus on broader understanding of customers through metrics
- Building consensus around metric definitions across teams
Personal Reflections
What struck me most was learning how data scientists dedicate significant time to “clarifying requirements.” This resonated with a conversation I had during a coffee chat with a retail sector data scientist, who shared that the most challenging and time-intensive aspect of their work is translating business problems into data problems.
The chapter covering fundamental statistical concepts like Simpson’s paradox and survivorship bias was particularly engaging, as it connected with my existing knowledge while adding practical context.
Deep Dive into Chapters 5-6: Practical Implementation
Key Learnings about Data Operations
-
Standardized Terminology
- Clear definition of terms like “data pipeline” across teams (the process of data collection, transformation, extraction, and utilization for decision-making)
- Importance of aligned understanding between data scientists, engineers, and marketers
-
A/B Testing Best Practices
- Managing confounding variables: A confounder affects the dependent variable without being an independent variable. Proper control of confounders significantly impacts experimental results
- Strict sampling criteria: Random sampling isn’t achieved simply by dividing subjects into odd/even numbers without careful consideration of confounders
- Evaluating A/B test value: The value of an A/B test should be judged not only by the p-value of the test results but also by considering the practical magnitude and impact of the experimental effects
Practical Applications
The book’s discussion of various services for building data pipelines has been particularly relevant to my personal projects, covering:
- Cloud environments
- ETL automation
- BI services
Looking Forward
This reading challenge has provided valuable insights into the practical aspects of data science in the private sector. As someone working on personal projects, I plan to:
- Revisit the book for deeper understanding
- Practice with the mentioned tools and resources
- Apply these growth hacking principles in real-world situations
A key takeaway: Just as UX designers need more than just the right mindset, becoming an effective data scientist requires both the mindset and the practical skills to implement data-driven decision-making processes.
Originally published as a three-part series on LinkedIn between November 8-20, 2024. This consolidated version aims to provide a comprehensive overview of my learning journey through “Growth Hacking” and its applications in data science.