A Few Takeaways From My First Job as a Data Analyst
Recently, I resigned from the first company that I worked at as a Data Analyst. It was rather a pleasant experienced but I resigned due to personal reasons that can’t be stated. Colleagues were fantastic, the organisation suited what I wanted as an ideal workplace, jobs were alright, no overtime and the list of praise goes on and on.
I would like to give credit to the CEO, Engineering manager and the one and only Head of Product for working well with me as well as being very accommodating and supportive towards me. I felt like we have known each other for a long time even though we have yet to meet each other (due to WFH regulation, of course).
Before sharing my job experience, I would like to share what I did on the day-to-day basis:
- Collaborate with respective stakeholders to identify specific product challenges
- Find trends and patterns within the data using the most appropriate techniques to deliver key insights
- Extract and gather the data required to solve the problems
- Perform data wrangling and cleaning
- Visualise data to make the user understand the current situation better
- Recommend data-driven strategies and actions to improve the products.
Enough blabbering about the company, here are some of the key takeaways that I learned during and after this job:
1. Effective Communication Is Key: Say whatever what you only need to say and listen to your coworkers well. Don’t bother wasting time in telling stories, especially you are in a flow-state and in working hours. Save the fun for later and spend time to do more research on delivering what the user needs.
2. Data Cleaning Takes The Highest Priority: When you have a raw and messy data, you will tend to deliver inaccurate representation of the data. This affected me pretty heavily when I was dealing with transaction amount. One huge outlier affected the average daily transaction amount, which in turn also affected the monthly transaction amount. Luckily I noticed the outlier early before reporting to the user.
3. Business Acumen Is Equally Important: If a Data Analyst doesn’t have a strong business acumen, all the data will feel like just numbers and letters, thus he or she acts like just data delivery guy. The Data Analyst can’t and won’t even give you new insights on the data collected, thus he or she said that everything performed as usual. No recommendation as on how to improve the product and no new initiative since he or she is not on the field to see how the business runs and have no idea what the end users really feel on the company itself. Thus, take sometimes to really understand deep enough about the business and see what the data really represents. Then only you can see what others can’t see.
4. Informative & Direct > Pretty Dashboards: No point in having colourful and full dashboards if those dashboards do not answer what your user needs. Your dashboard should have only the necessary information and aesthetics so that you can provide the most straightforward answer to what your user question. If there is no need for a feature (i.e. multiple colours, doughnut chart, violin plot, etc), don’t include it. It will cause the user extra confusion and will take your precious time to answer his/her questions and remove the feature at the end of the day.
5. SQL is your main weapon of choice: Master your SQL well since you will be dwelling most of your time there, whether extracting or manipulating the data that we require. You should master the basic statements such as SELECT, CREATE, DELETE and etc. Besides, you should learn how to deal large data in SQL so that you can have the shortest time and optimised statements which make you an efficient Analyst. Those are the things you should know by heart.
6. Take your time, but not too long: When doing data analysis, we should learn how to take your time on tackling the tasks that are given to us. By taking your time, we can do in depth research, thus giving more thorough insights that can help to solve the problem. We should slowly dissect what our user wants carefully and translate the problem statement into the deliverable that our user desire and understand best.
That’s all that I have. Hope it can give your more insights on what Data Analytics is. You can find me through LinkedIn here! Thank you for reading this! Sorry, no visualisation is included since I do a lot of visualisations on my job. Text is enough to deliver my message :p