Early Bird: Save 10% when you enrol before 30th April.
Early Bird: Save 10% when you enrol before 30th April.
Data Analytics
How to get started in Data Analysis - All your options explained
Written by lecturer Mark James
Data analysis has become a critical capability in modern business and society. The tools you decide to use for analysing data can influence the outcomes as much as the data itself. The main options are traditional spreadsheet software like Excel, Business Intelligence platforms such as PowerBI and Tableau, and programming languages such as Python and SQL. Each tool has its own unique strengths and limitations. And the choice of whether to go the code or no-code route is not as straightforward as it once was.
)
Excel - The Accessible Powerhouse
Excel is often the first tool many people encounter when dealing with data, and for good reason. Its interface is intuitive, using a grid-based layout that mirrors the old-style paper spreadsheets. The learning curve is pretty low, making it an excellent entry point for beginners, and you can scale up to a pretty advanced level as needed. With its built-in functions, Excel is powerful enough to handle a broad array of data analysis tasks, from simple calculations to statistical analyses and even some machine learning. With the coming introduction of the AI-powered Microsoft 365 Copilot before the end of the year, Excel's value and accessibility is likely to increase. Copilot uses natural language processing to simplify complex data analyses, making it an ideal choice for people who don't work with large, complex data sets. Overall, it is an excellent tool that is often used in combination with others as part of the tool stack for many data analysts.
Microsoft Power BI - Advanced Analytics without Coding
Power BI operates in the space between Excel's user-friendly interface and the more robust capabilities of specialised data tools. It's particularly strong at connecting disparate data sets and transforming them into interactive dashboards. Power BI uses a drag-and-drop interface that's accessible to users who may not have coding expertise but still want to perform advanced analytics. However, as Power BI is a Microsoft product, it is often best suited for environments that are already invested in the Microsoft ecosystem. And while PowerBI offers impressive data visualisation capabilities, it lacks the granularity and customisability of code-based approaches. Therefore, the tool is ideal for analysts who need to combine data from different sources and create compelling visual narratives but don't need the raw power or customisation that coding offers. Also, unlike Tableau, the free version of PowerBI is fully functional - minus the cloud integration - which is nice!
Tableau - Visual Storytelling
Like PowerBI, Tableau is highly optimised for dealing with large data sets. Its strength lies in its self-service visual analysis features, which empower users to explore data interactively and ask ad-hoc questions. But unlike Power BI, Tableau is platform-agnostic – it can run on OS’s other than Windows, and it integrates well with various databases and data sources. However, Tableau's learning curve is steeper, and its pricing model can be prohibitive – it has a free version, like PowerBI, but with significant limitations. Its specialisation in visualisation also means it's slightly less well-suited for other kinds of data manipulation or computational tasks. Tableau is really for those who are heavily invested in data analytics, need to handle big data, and want the freedom to explore data without pre-defined questions. For the most part, PowerBI and Tableau offer the same powerful capability, but with small variations in different aspects.
SQL and Databases
SQL, which stands for Structured Query Language, is a programming language designed for tasks such as managing and manipulating relational databases. It's a commonly used tool. SQL offers a powerful, flexible way to interact with data, allowing users to perform complex queries, joins, and aggregations with ease. Unlike the other tools I've covered so far, SQL requires coding skills, making it less accessible to the general user. However, its language syntax is pretty straightforward. SQL is the tool of choice for those who need to perform complex data manipulations and are comfortable with coding. It often serves as the backbone for more advanced analytics pipelines, feeding data into the likes of Power BI or Tableau for visualisation.
Python - Power and Flexibility
Python is an incredibly versatile programming language with a rich ecosystem of libraries designed for data analysis, such as Pandas and NumPy, among others. While Python requires a steeper learning curve than Excel or Power BI, its power and flexibility are unparalleled. Python scripts can automate repetitive tasks, handle large data sets, and perform complex computations. It's highly customisable and can be integrated into almost any data pipeline. Python also supports advanced statistical models and machine learning, opening up avenues for predictive analytics. However, Python's power comes at the cost of accessibility. Its code-based nature means that it's not suitable for users who are uncomfortable with programming. For those who don’t mind the learning curve, Python offers an almost limitless array of possibilities for data analysis.
Excel and Python
Some of the most exciting developments in this field are occurring at the intersection of coding and non-coding environments. Microsoft have just announced the integration of Python into Excel, bringing together the accessibility and familiarity of Excel with the power and versatility of Python. Users can now use Python's advanced libraries like Pandas and Matplotlib within Excel's user-friendly interface. Python calculations in Excel are cloud-based, courtesy of Microsoft's Azure, and therefore make the power of your computer/laptop effectively irrelevant. It will be particularly appealing to people who are dealing with huge datasets but are hesitant to move fully to specialised, code-based tools. It really opens up the power of these tools to a broader audience - people who are proficient in Excel but not necessarily skilled in programming - to be able to engage in more sophisticated data analysis. In addition to AI capabilities like Copilot, Excel is becoming a more adaptable, robust platform for comprehensive data analytics.
Choosing Between the Coding and Non-Coding Options
The decision to go the coding or non-coding route is increasingly becoming less of a 'one or the other' and more of a 'where should I start?'. The integration of Python into Excel illustrates this. People who might be hesitant to commit fully to specialised, code-centric tools can now access advanced analytics capabilities in Excel using Python or combine with Power BI or Tableau. These hybrid solutions are blurring the lines between traditional choices, offering a compromise between user-friendliness, power, and flexibility.
If you do decide to avoid coding altogether, non-coding tools like Excel, Power BI, and Tableau offer plenty of capabilities, with the advantages of accessibility and intuitive interfaces. On the other hand, coding provides unparalleled flexibility and scalability, but requires a deeper skill set and is often less accessible to the general user. Your choice will depend on various factors, including the size and complexity of your datasets, the specific analytics tasks you need to perform, and your comfort level with coding. Each tool has its own strengths, limitations, and ideal use-cases. The key is to find the tool - or combination of tools - that aligns best with both the task at hand and your own skill set. As technologies continue to integrate and evolve, expect to see even more options that offer the best of both worlds.