Rowley’s story from chemical equations to data Solutions

I've always been a numbers-oriented person—regardless of what I'm doing, I find myself drawn to the underlying figures. My background in Chemistry at University meant I used Maths heavily throughout the course, which I believe developed a strong analytical mindset. This mindset was further developed during my time working in medical communications. Working in a highly-regulated industry such as UK pharma, you need a keen eye for detail and an in-depth understanding of the code of practice.

Throughout my previous job roles, I endeavoured to incorporate data wherever possible. I thoroughly enjoy collecting personal data, whether it's through Spotify Wrapped or my smartwatch, and I tend to gravitate towards media related to data analysis. I'm particularly fascinated by the algorithms used on platforms like YouTube and Amazon, and how they generate recommendations based on one's history with varying degrees of accuracy.

Instead of watching popular TV shows and movies, I largely watch sports data analysis videos on YouTube. I've always been captivated by the stories told and the data which supports these narratives (example YouTube videos: The History of the Seattle Mariners, The History of the Atlanta Falcons, Where is the toughest place to play on a Cold, Wet Tuesday night?)

I found the course through talks with a work coach. I had mentioned to them that I wished for a change in career path, and they showed me possible courses I could join. I noticed that this one was due to start soon, and so I had a "now or never" moment and decided to jump in.

I didn't have any goals at the beginning of the course; my only goal was to do "as much as possible". I wanted to learn SQL, Python and R before the end of the course; however, this changed to SQL, Python and Tableau as it more closely aligned with what I would want to do in a professional role and what I thought had the most practical use.

I already understood statistics from my experiences of maths and science, but I had no means of applying this to data as I didn't know how to query. I wanted to "scale-up" my ability to apply statistics to data, whether that be increasing the size of the datasets used, the complexity of the data, or the accuracy of my analysis.

The Bootcamp Experience

The format of the bootcamp is largely hands-on work. There is usually a presentation about underlying concepts and context behind what we are studying, and then we would have a dataset with example questions where we can apply the new knowledge. If anyone is struggling, we can ask peers for help or Matt goes through the correct solution to keep everyone on pace. The size of the dataset increases each week, which demonstrates your expanding abilities, and towards the end of the course, you become more independent with sourcing your own data for a project.

I learned everything to do with SQL, Python and Tableau. Also, a background on human cognition/perception and how to present data to improve understanding and enhance narrative and reduce cognitive load. I found this incredibly insightful, as I had identified the succinct communication of insights as a potential weakness - it is one thing to understand your data, but getting someone else to understand it is a different task.

Personal Projects and Achievements

I'm currently working on my fitness data, from the past 5 years of wearing Fitbit and Garmin technology. This has been a goal of mine since I bought my first watch; I knew I wanted to analyse data but I didn't know how to, so I just settled on collecting data until I eventually could. Secondary reasons are that I wanted to practice working with "dirty" data, and given that Fitbit and Garmin store their data in different ways, joining this data would require significant cleaning first. Finally, I wanted to really focus on the presentation of data, as I had identified this as one of my weaknesses. I wanted to produce an aesthetically pleasing piece which was understandable, with a clear message or insight.

I'm proud of my holistic improvement in my approach to data analysis; I have uploaded workbooks to Tableau Public so I can see my progress over time. I have immense pride in seeing this development, and I extrapolate this to "how good can this be if I did it in a year's time, or 5 years". My most recent work on my fitness data is my current favourite as I was able to implement highly complex Tableau features to create a bar chart with a spiral axis. This is only one aspect of a larger project on my fitness, relating to daily steps; however, I am really proud of how this turned out.

Some of the biggest challenges I faced in the bootcamp was the visualisation of data - identifying what the audience needs to know, what they already know and how do I tailor the presentation of my data accordingly. The Gestalt Principles and pre-attentive processing classes/lectures were super useful in helping me tailor my visualisations to this.

Matt talks about superpowers and I think the "lateral thinking" aspect is what I excel at, finding alternative routes to query data when one route is unsuccessful. For example, I wanted to visualise the history of traffic collisions over the last 5 years, which included data from 4 separate tables, the largest of which had over 1 million rows of data. Any query that I used required significant processing times and rendered any analysis painstakingly slow. There were also issues when joining the data causing an exponential increase in the number of rows. Therefore, I had to find ways of speeding up this process while also eliminating the joining errors.

I learned that the order and structure of queries was incredibly important: splitting one long query into multiple shorter queries, keeping data as simple as possible until you need to convert it into something insightful. For example, there were many "types of vehicle" in this dataset, but stored as numerical values with definitions in another table (1 = car, 2 = motorbike, etc). Querying the data as numerical values first significantly reduces processing time. I used lots of subqueries to merge the definitions list, which also impacts processing time, so if I were to do this again, I may try to find a way to circumvent using so many subqueries.

My biggest takeaways from the course are:

"Just because you can do something, doesn't mean you should". This also relates to the "Dashboard Real-Estate" being valuable so only keeping things that justify/enhance your narrative.

"Funneling down" to create a cohesive story.

Focusing the story that you are trying to tell to one or two points and using the data to reinforce this throughout your story. Too many points may obfuscate the overall message you're trying to portray.

Concentrating on the audience. Presenting data to your peers will be different to presenting to a non-technical audience.

Querying languages are just like any other language - there are often many different ways to "say" the same thing so you should identify what works best for you/your system. However, just as you find one method works best for you, this may not be true for somebody else, so you need to have that well-rounded understanding of the languages so that you can translate other people's work.

Learning Environment and Support

I love the training venue at The Curious Academy. If you ask most people to picture a "data analysis venue", they would most likely think of something impersonal, maybe boring, antisocial, but the CA is anything but. The décor, the team, the café, the open office where professionals work, it all humanises data analysis. Everyone is warm, welcoming and seem willing to go above and beyond to help you succeed. The people make every class a pleasure to attend.

Equipping me with the tools to analyse the data has been only part of the process. The bootcamp instilled confidence with gradual increases of difficulty, and fostered a supportive environment where it's okay to make mistakes. There is a constant stream of support through the course, especially with employability towards the end of the course. The talks with real-world data analysts/scientists help to personalise the profession and make it less daunting. The Data Club allows you to meet people further into the course, so you can learn from your peers, or conversely it provides you an opportunity to share your knowledge with more recent newcomers. A lot of my concerns are largely down to self-confidence, and the team has done nothing but push me to believe in my own abilities.

Future Aspirations

I don't have a specific sector in mind when looking for a new career; I often think about the "greater good". To be generic, I am a team player by all accounts and I want to be in a role that creates good, meaningful change for people. This change could be for my colleagues, our customers, or the general public. My philosophy is to "help the most people, as much as possible". If I am in a team, I want our team to succeed as a whole.

Given how my interest in analysis was cultivated from online media, I would love to develop my skills in data communication. While I have no concerns about my confidence in presenting, I believe that one of my biggest weaknesses is conciseness. I think that if I create scripts when explaining my data, it would afford me the opportunity to keep on-track when delivering a narrative.

I'd like to continue improving my knowledge of Python, as it has so many applications if you're able to understand it. Further to this, I think that analysis of unstructured data is an area that I would like to explore - analysing images, text strings. And while it may be far down the road, I would like to develop an understanding of machine learning and AI.

Advice for Future Students

If I were to give advice to someone considering attending a similar bootcamp I would tell them that the output will match your input. The amount of work you put in will correlate with how much you learn/achieve. I have a couple of "learning" tricks that I stick by:

The forgetting curve: If you learn something once, you will remember less than 50% within the hour. Within a week, you will remember less than 25%. Therefore I used a lot of repetition when learning to constantly reinforce that knowledge. DataCamp Practice is good for this, but also replaying courses that you have already completed is really useful.

Cognitive overload: It's really tempting to throw yourself at everything and getting as much done as possible. But this most likely is making your learning less efficient as your working memory can only handle so much new information before getting overloaded. To learn effectively, you should try to focus on one topic, and ensuring you understand that before moving to a new topic. I would split my mornings and afternoons into different subject areas, with a break in between, and try to remove as many distractions as possible.

Diversify your learning style: There is only so much information you can gain from watching the same video over and over again. Change it up, do some practical work, or write notes on the video, do what you think works for you. If you change the medium of your learning, then you are giving your brain more opportunities to develop/solidify those pathways. Once again, DataCamp does have a diverse range of exercises, so give them a try and don't be disheartened if you're struggling.

Learn how you work best: I used to write notes for everything, but I generally dislike reading as a method for learning so would never re-read my own notes, rendering the exercise relatively pointless. Instead, I find that I "learn by doing" best. So I did a lot more practical work, and tests for understanding. Discussing work with my peers also helps, not only can they teach you something, but I find my understanding improves when explaining something to someone else so it is mutually beneficial. Other people may have different strengths to you, but you're all on the same journey so collaborate and teach each other.

The most exciting aspect for me is the possibilities ahead. It's a new chapter for me, and I'm approaching it with a completely open mind. I have no idea of where I'll end up but the journey that I am on now fills me with positivity. I hope that my next role will allow me to build upon the foundation the course created, while also providing opportunities for further learning.