• Learn the fundamentals of Statistics
• Learn SQL
• Learn Python for Data Analysis
• Learn Data Manipulation and Visualisation
• Learn Statistical Analysis
• Learn Data Visualisation Tools
• Work on Projects
• Learn Data Storytelling

Introduction

The last post in this Data Analyst Roadmap series was regarding the use of Data Visualisation tools. Given that I have previously outlined how I have displayed understanding and analysis within statistics, SQL, Python, and data manipulation & visualisation, it is now time to use all of these tools and techniques in projects. This seventh step, the penultimate step, displays the application of theoretical knowledge through the gateway of practical application, problem-solving prowess, and portfolio enrichment.

I thought today would be a fantastic opportunity to reflect on some projects that I have developed and explore what makes them unique. Each project not only encapsulates the skills acquired along the roadmap but also embodies the spirit of creativity, exploration, and learning. For each project, I’m going to outline:

• Overview,
• Favourite Aspect,
• Key Learnings, and
• Areas for Improvement.

Project 1 – Powerade

Using brand sales-generated data, I wanted to develop a Dashboard that provided an overview of sales information for management.

Favourite Aspect: Dynamic visualisation of sales data, providing actionable insights for retailers. Even though I understand how all data can be linked between data sources and visualisations, it still blows my mind how changing on cell can have flow-on effects throughout the entire page.

Key Learnings: Enhanced understanding of data-driven decision-making in the retail sector, proficiency in dashboard creation. Throughout my road journey of data analytics, I knew that my biggest personal pain point would be the visualisations. Here, I wanted to continue honing my craft around visual concision. If I can use less space, and fewer images, but still provide a huge impact, then I have achieved my job for management. It is a craft that I will be working on in perpetuity as each project and person will have different requirements and styles.

Areas for Improvement: Further refinement of user interactivity features. As mentioned in the post, Google Sheets currently has a limitation around Pivot Chart slices, especially across sheets. I would also like to include forecasting measures and provide machine learning elements to narrow the model’s accuracy and improve the forecasting probability.

Project 2 – Data Professional Survey

Using a Data Profession survey, I wanted to manipulate qualitative data in a quantitative manner to show an overview of a subset of the profession’s participants and how they feel within it.

Favourite Aspect: Conducting comprehensive data analysis to extract meaningful trends and insights from survey responses. Given the content of this survey, it was important for me to explore, observe, and show the link between industry sector and individual satisfaction for data professionals.

Key Learnings: Survey data processing, statistical analysis techniques for survey data. Specifically, taking qualitative data and transforming it into a quantitative way. The PowerBI post highlights how I was able to do this.

Areas for Improvement: Incorporation of advanced sentiment analysis techniques. By allowing users to show magnitude feelings within the ‘select 1 to 5’ method of answering questions would allow for a greater depth of meaningful insights and trends into survey responses.

Project 3 – Weather Dashboard

Dashboard of live weather data, with forecasting view.

Favourite Aspect: The user enter any city (even vaguely) into the dashboard and the data will be retrieved instantaneously.

Key Learnings: How to utilise API (.csv or .tsv) data and image import functions to display a dynamic spreadsheet.

Areas for Improvement: Integrate more accurate weather conditions forecasting by updating current weather and comparing accuracies (along with its changes) over time.

Project 4 – Analysing Outliers with Z-Scores

Using confectionary sales-generated data, I wanted to develop a Dashboard that provided a mix of marketing and sales pipeline information management.

Favourite Aspect: Uncovering hidden patterns and anomalies in sales data using statistical methods. Taking this post from step 1, to step 2, and (this) step 3, I thoroughly enjoyed the practical application of econometric analysis on a sales dataset. By identifying anomalies, removing them, and adjusting that more accurate information to a business, allowed me to introduce efficiency and accuracy in budgeting, analysis, and forecasting.

Key Learnings: How to take a sales dataset, apply allocation of Z-scores and identify outliers, and accurately interpret those outliers for analytical results.

Areas for Improvement: Exploration of alternative outlier detection algorithms for comparison, and incorporating these algorithms into a Seasonal Auto-Regressive Integrated Moving Average (SARIMA) model for forecasting sales.

Project 5 – Uber

Using US-based rideshare information, I was interested in exploring Google-based cloud services to create an end-to-end data analysis pipeline.

Favourite Aspect: Even though I want to say that my favourite aspect was analysing ride-sharing data to understand user behaviour and demand patterns, I think it was actually learning a new environment, specifically a Google-based cloud platform.

Key Learnings: Learning the go to woah of Google cloud, for Data Analysis, allowed me to stably handle real-time data streams and geo-spatial analysis for location-based services. What was helpful in this learning was mapping out my proposed system and integrating each component within that pipeline.

Areas for Improvement: Integration of additional data sources for comprehensive analysis.

Conclusion

As we reflect on these projects, we see the culmination of our journey through the Data Analyst roadmap. Each endeavour has been a stepping stone, honing our skills, expanding our horizons, and fueling our passion for data analysis. Looking ahead, the final step beckons – Data Storytelling, where we’ll learn to weave narratives from data, bringing insights to life. But for now, let’s celebrate the achievements, embrace the lessons learned, and embark on the next phase of our data adventure with renewed zeal and determination.