Become a Business Expert

This is a common piece of advice, and for good reason. Grasping the nuances of the business you’re part of is crucial for extracting value from the data collected. The more knowledge you acquire, the simpler it becomes to generate positive analytical outcomes.

How can you deepen your understanding of your business?

• Engage in Conversations: We will delve into this aspect later in the article, but initiating discussions with anyone willing to share their role and its impact on revenue is vital.
• Investigate Customer Support: Pay attention to customer feedback, product usage, and the reasons behind customer dissatisfaction. If possible, experience the product as a customer yourself.
• Analyze Competitors: Explore whether competitors share similar pricing strategies or team structures. Identify key differentiators between your company and theirs.

Once you have a solid grasp of the business landscape, delve into theoretical knowledge.

What factors contribute to the success of a B2B company? What strategic decisions must be made (e.g., "quality versus quantity")? What various revenue strategies exist in your sector? What do ARPU, LTV, and SAC mean?

> Data scientists and analysts typically aren't responsible for reporting financial metrics like billings, MRR, or EBITDA, which fall under finance and accounting. Instead, we focus on transactions, subscriptions, and funnel conversions, often facing challenges in aligning total subscriptions with ARR or new transactions with net revenue. This requires nuanced understanding and context. (1)

The theoretical aspect is often overshadowed by practical business narratives presented by stakeholders and colleagues. However, this foundational knowledge is essential. Learning business theory, whether B2B or B2C, exposes various concepts that may challenge your current business understanding.

While not all theoretical knowledge will be directly applicable, it will enhance your problem-solving skills and adaptability to specific scenarios and shifts in the business environment. Embracing this challenge will demonstrate your maturity and ability to handle complex situations, earning the trust of stakeholders.

A well-informed data analyst who surpasses even the stakeholders' expertise can be a tremendous asset to any organization. Strive to be that individual.

Gain Insight into Data Engineering

From the perspective of a data analyst, data engineering is a more technical discipline. While data analysts should possess technical skills, they need not delve too deeply into data engineering tasks.

Top analysts understand the historical context and foundational architecture concepts of data engineering, including:

• Big Data Evolution: Analysts depend on upstream pipelines to obtain data. These pipelines often utilize significant machinery for data transformation and aggregation. "Big Data" has been evolving for two decades, and it’s vital to comprehend concepts like Map Reduce patterns, distributed computing, and cloud technologies.

• Architectural Frameworks:

  • Online Transactional Processing (OLTP): This pertains to databases utilized by customer-facing applications, optimized for rapid processing and CRUD operations.
  • Online Analytical Processing (OLAP): OLAP supports dashboards and business intelligence by efficiently processing large data volumes and enabling complex aggregations and filters.
  • Batch vs. Streaming: Batch processing involves handling a large volume of files simultaneously, suitable for analytics where reports don’t need immediate attention. Streaming processes each file as it arrives, ideal for scenarios demanding real-time data.

Learn Python, SQL, and BI as Code

Numerous programming languages exist, but within the data realm, two dominate: Python and SQL.

• SQL: Data is frequently stored in databases or data warehouses. SQL, a declarative language, simplifies querying data and is essential for serious data analysts.

> In addition to SQL, mastering frameworks like dbt or SQLMesh can be beneficial. These tools help manage dependencies between SQL queries at scale, particularly when multiple teams utilize a data warehouse for analytics.

• Python: Known for its ease of learning, Python has been the preferred language for data scientists and engineers for years. It excels in writing business logic and data manipulation.

Mastering these two languages will equip you for professional data analysis roles, as they are standard in every organization.

However, conducting comprehensive analyses can be challenging, as neither SQL nor Python is tailored for deep analytics. An emerging trend in analytics is the concept of "BI as Code," supported by tools like Evidence, Malloy, and Rill, which aim to streamline and enhance analytical processes.

Proficiency in this area can elevate you to the top 1% of analysts, enabling you to produce clean, repeatable, and in-depth analyses while saving significant time.

Engage Beyond Casual Conversations

Interactions with stakeholders and engineers are often undervalued.

• Engage with Stakeholders: Business decisions frequently rely more on intuition than on precise data. Understanding individuals’ data literacy and personal traits will help shape your presentations.
• Connect with Engineers: Remember, data serves as a representation of reality, not reality itself. Understanding the trade-offs made by data engineers in the upstream processes and data collection methods can reveal critical bottlenecks and biases, which could skew your analysis.

By engaging with everyone involved with data—and indeed, anyone in the organization—you will gain insights into both people and data, the two crucial elements of any modern enterprise. Make this a proactive part of your routine.

When starting a new role, prioritize scheduling short meetings with as many colleagues as possible. This initiative sends a strong message about your motivation and character, and can reignite creativity in those who may feel stuck in internal challenges.

Master Storytelling and Human Cognition

Whenever we analyze data, we:

1. Filter the information.
2. Search for meaning.
3. Act based on previous experiences and inherent biases.

This process is particularly pronounced for business stakeholders, who often have limited time and data literacy.

To enhance the results of data analysis, it’s crucial to recognize the biases and heuristics influencing these stages.

As noted earlier, initiating conversations is vital. It will help you map out everyone’s perspectives and concerns.

The real challenge lies in presenting your findings engagingly. Your audience should feel compelled to act based on your insights, making your efforts genuinely valuable. This requires skills in storytelling, data visualization, graphic design, and an understanding of human cognition.

Conclusion

Here’s a quick reference guide:

• Become a Business Expert: Absorb as much knowledge as possible about your organization. Engage with everyone, from customer support to executives, to grasp both the big picture and intricate details.
• Understand Data Engineering Basics: Familiarize yourself with data pipelines and collection processes to avoid misinterpretation in your analyses.
• Master Your Analytical Toolkit: Python, SQL, and BI tools like Malloy are essential for data wrangling and insight generation.
• Develop Conversational Skills: Don’t underestimate the importance of dialogue. Engage with stakeholders to identify their needs and biases, and connect with engineers to understand the data's journey, enhancing your analysis's impact.
• Cultivate Storytelling Abilities: Data carries immense power, but it needs a compelling narrative to resonate. Sharpen your storytelling skills while considering data visualization and human psychology.

By following these strategies, you can evolve from a data analyst into a strategic partner, the kind of data translator every CEO dreams of having on their team. Remember, data is just the beginning; your ability to turn it into actionable insights is what truly adds value.

If you're seeking valuable data resources and insights, consider subscribing to my newsletter — From An Engineer's Perspective.