Becoming a data hero: a non-programmer’s journey to mastering data

Introduction

The world is becoming increasingly data driven. Not so long ago, most decisions were made based on experience and gut feeling. Today, professionals across various fields need to analyze and handle data to make informed choices. For some, this shift is seamless – they live and breathe data, effortlessly extracting insights from complex datasets. But for many non-programmers, this new reality can be daunting.

Imagine navigating this data-driven landscape with confidence, without needing to become a full-time programmer. That’s the promise of becoming a “data hero.” A data hero is someone who, regardless of their programming background, can effectively handle data to gain better insights and automate tasks, thereby excelling in their profession.

This article is inspired by the Data Handbook ([Lijd24]), a guide designed to help non-programmers master the essentials of dealing with data. The handbook emphasizes that you don’t need to know everything about programming to deal with data effectively. Instead, by focusing on seven core data skills, anyone can transform how they work with data.

In the following sections, I will explore these seven essentials, discuss how to develop a data-centric mindset, and provide practical steps to become a data hero in your field.

The rise of the data hero

The ability to handle and interpret data is not just a nice-to-have skill but a necessity. Professionals who can navigate the complexities of data are highly in demand. But what if you’re not a programmer? The good news is that you don’t have to be one to become a data hero. By mastering certain core skills, non-programmers can effectively handle data, automate tasks, and make better-informed decisions.

Why non-programmers need data skills

Data is everywhere. Whether you’re in finance, marketing, healthcare, or any other field, chances are you deal with data regularly. Traditionally, non-programmers have relied on tools like Excel to manage data. However, as datasets grow larger and more complex, these tools can become limiting. Moreover, the demand for data-driven decision-making requires a deeper understanding of dealing with data techniques.

By acquiring essential data skills, non-programmers can:

• improve efficiency: automate repetitive tasks and reduce manual errors;
• gain better insights: analyze data more effectively to inform strategic decisions;
• enhance collaboration: communicate more effectively with technical teams.

The seven essentials of dealing with data

The Data Handbook introduces seven core skills – referred to as “Essentials” – that form the foundation for dealing with data. These Essentials are universal and apply across all tools, whether you’re using Excel, Python, SQL, Alteryx, or any other dealing with data tool. The key is understanding that while the tools may differ, the fundamental steps in solving data problems remain the same.

1. File input. The first step in any data project is importing data from various sources such as CSV files, Excel spreadsheets, databases, or APIs. Understanding how to read data into your chosen tool is crucial. The method you use will depend on your tool’s capabilities and the format of your data.
2. Filtering data. Filtering allows you to focus on specific subsets of your data by selecting rows that meet certain criteria or excluding irrelevant information. This step helps you deconstruct your dataset into more manageable parts, making it easier to work with.
3. Grouping data. Grouping involves aggregating data based on specific categories or keys, such as summarizing sales data by region or department. This helps you analyze data at different levels of detail, essential for gaining meaningful insights.
4. Merging and joining data. Often, the data you need is spread across multiple sources. Merging and joining techniques enable you to combine datasets based on common keys. This step reconstructs your data into a comprehensive dataset that can answer more complex questions.
5. Transforming data to long format. Data can be structured in wide or long formats. Transforming data to a long format is useful for certain types of analyses and visualizations. This process involves deconstructing wide tables into a standardized format where each row represents a single observation.
6. Transforming data to wide format. Conversely, transforming data to a wide format is necessary for other types of analyses. This involves reconstructing your data so that multiple observations are spread across columns, which can be essential for specific modeling techniques or reporting requirements.
7. File output. Finally, you need to export your processed data. This could involve saving it to a new file format, generating reports, or loading it into another system. This step completes the reconstruction phase, providing you with the final product that addresses your initial data problem.

Developing the “Deconstruct to Reconstruct” mindset

Mastering the Essentials is not just about learning specific functions or commands in a tool; it’s about adopting a new way of thinking – the “Deconstruct to Reconstruct” mindset. This approach involves breaking down complex data problems into smaller, manageable parts (deconstructing) and then systematically solving each part to build up your final solution (reconstructing).

Choosing the right tools

With numerous data tools available, selecting the right one can be overwhelming. The key is to choose a tool that fits your specific needs and circumstances. The Essentials remain the same across tools, but factors like data size, complexity, and organizational constraints may influence your choice.

• Excel: great for small datasets and quick analyses. It’s user-friendly but limited in handling large datasets.
• Python and R: powerful programming languages for data analysis. They require some programming knowledge but offer greater flexibility and can handle large datasets efficiently.
• Alteryx: an easy-to-learn, easy-to-use tool for data analysis with a graphical interface. It’s excellent for those who prefer visual workflows but is limited to the functionalities present within the tool.
• SQL: ideal for working directly with databases. It’s useful for extracting and manipulating large datasets without needing to export data to another tool.

Deconstruct: analytical thinking

Begin by understanding the most granular elements of your data – the smallest units that make up your dataset. Ask yourself:

• What is the smallest unit I can analyze?
• Can I obtain more detailed data from the existing dataset?

By focusing on these granular elements, you can deconstruct your data problem into fundamental building blocks. This analytical thinking allows you to identify exactly what data you need and how to manipulate it to gain insights.

Reconstructing data

After deconstructing your data into its basic components, you can reconstruct it to solve your specific problem. This involves using the Essentials – filtering, grouping, merging, and transforming data formats – to piece together the solution. By addressing one building block at a time, you can systematically build up to the final answer.

Practical example

Imagine you have a spreadsheet of customer orders that’s not structured for analysis. By deconstructing it, you recognize that each row represents an individual order, with details like date, product, and quantity. Using the Essentials, you can filter orders from the last quarter, group them by product category to view totals, and merge this with pricing data to calculate revenue. Reconstructing the data in this way turns a messy spreadsheet into a clear sales report, demonstrating the power of this mindset.

Learning to ask the right questions

Deconstructing and reconstructing your data is accomplished by applying the Essentials. If you don’t know how to apply a particular Essential, you can find guidance on the internet. In the digital age, answers to almost any question are readily available – if you know how to ask. Formulating the right questions therefore becomes crucial. Two key concepts for asking the right questions are:

• Use proper terminology: familiarize yourself with data jargon and the specific terms used in your chosen tool. This makes your searches more effective and helps you understand the solutions provided.
• Be specific: instead of asking, “Why isn’t my data loading?” consider, “Why does ‘read_csv’ in Python return an encoding error when reading a UTF-8 file?” Specific questions yield more precise answers.

By combining the universal Essentials with the “deconstruct to reconstruct” mindset, and by asking the right questions to search engines and AI assistants, you will be equipped with all the ingredients required to tackle any data problem, regardless of the tool you use. The Data Handbook serves as a guide to help you ask the right questions, empowering you to become a data hero in your field.

Conclusion

The data-driven world isn’t reserved for programmers alone. By mastering the seven essentials of dealing with data and adopting a “deconstruct to reconstruct” mindset, non-programmers can become data heroes in their fields. Whether you’re an accountant, marketer, or healthcare professional, these skills will empower you to work more efficiently, make better decisions, and advance in your career.

The journey to becoming a data hero starts with a single step. With resources like the Data Handbook and by asking the right questions, anyone can unlock the power of data.