Top 4 Skills A Data Analyst Must Have
The world of data is fast growing that more than half of the entire data we have for analysis today actually come from the last 5years. Think about what data would be like in the next 10years.
Data Analytics and Data Science have been touted as the “sexiest job”. Some use the tagline “Data is the new oil”. Although majority of us in the industry don’t buy the later concept, however, we do agree that of all the tech related branch, only data remains when all is done and gone.
Programming languages come and go, frameworks reign and extinct. While they are still active, they keep leaving us with “Data”. This is one of the reasons we know that this industry is here to stay. Another reason I believe it is here to stay is simply because it is not an industry specific career. As a matter of fact, it is needed by all industry.
In my own opinion, the people that make the best data analysts are those coming from their industry to start a career in Data Analytics. The reason is simple, it is easier to get carried away if you are a more generic data analyst, but if you are an industry specific data analyst, it helps you settle in quickly with data specific to your industry.
For instance, if you are a Medical Lab Technician, when you settle as a data analyst in the health sector, you should do better than a generic analyst. This is because of your “Domain Knowledge”. We shall discuss this further as the 5th skillset for s data analyst.
Now let’s dive deep into the main discussion for the day. A data analyst identifies a problem, gets relevant data, cleans the data, then transform, visualize and model it to discover meaningful and useful insights that will help solve the problem. I will not bug you with what SQL, Excel, Tableau or Power BI is. You can easily look them up on Google.
1. Structure Query Language (SQL)
Let’s say there is a gender imbalance in your company as it is the case in the tech industry, the government recently passed a law that mandates all companies in the country to give equal opportunity to both male and female employees. As an analyst, you mostly will be the person to get these data where there is no Data Engineer. To get this data, you need to query the database(s) where these employees information are stored, this process is known as EDA (EDA stands for Exploratory Data Analysis). This is where the knowledge of Structure Query Language becomes useful. Most companies store their data in sql databases, hence the reason you need to go for sql over no-sql. The flow is to simply learn the standard SQL using any of the sql servers (Mysql preferably), the other server have very little or no difference. The 4 most popular sql servers are Oracle, Mysql, Microsoft SQL and Postgresql.
2. The Power of Tableau and Power BI For ETL
The second stage is to export this to a csv or an excel file (csv stands for comma separated values) for cleaning and transformation (note: you can also export to other file formats like the .sql). You can also do this cleaning in the sql server, however, I will advice you do your cleaning in advance tools like Tableau or Microsoft Power BI. (BI stands for Business Intelligence). This tools mostly do not require coding, as their user interfaces are very intuitive. This is where you spend more than 70% of your time. The quality of your data heavily depends on this stage, hence you must be very patient and careful. This process is what is popularly known as ETL (ETL stands for Extract, Transform and Load) We shall talk about cleaning in another article.
3. Tableau and Power BI for Visualization and Modelling
The next move is the visualization. The beautiful thing about Tableau and Power BI is that you can do the ETL, visualization and modelling in them. They are the best in the industry right now. So a very sound knowledge of both of these tools will give you a competitive advantage. After getting your insights, you build your dashboard. A dashboard, in my understanding is a collection of visualizations on one page for proper storytelling and decision making.
4. The All In One Microsoft Excel
Now one tool I have not actually explained is the excel spread sheets. This is by far the more used tool by analyst in general for small dataset. Yes, where your dataset has like ones and tens of thousands of rows, using excel for your ETL, and visualization becomes very handy. Now as you know, I used the term Microsoft Power BI, this means both Power BI and Excel are owned by Microsoft. I will say Power BI is an upgrade to Excel. If you know Excel, then you should comfortably pick up Power BI. Excel does everything that Tableau and Power BI does but for smaller dataset. The caveat here is that, Tableau and Power BI are built with more power and intuitive user interfaces.
5. Domain Knowledge (Bonus)
lastly, let’s talk about the least spoken yet super important skill. That is Domain Knowledge. This is simply having a knowledge of an industry where you can apply your data analytics skill to. If you are a petroleum Engineer, you should do better as a data analyst in the oil and gas sector than in the sports sector. I am personally interested in the Education and Financial sector as I have very good understanding of their terminologies.
To conclude this article, I will advice you to first go for SQL, then Tableau, followed by Excel and finally Power BI. When you are comfortable with all these tools, you can then gradually start your programming journey with Python or R which are the most popular in the industry. Remember, anything you can do with these programming languages as a data analyst is far more easier with Tableau and Power BI.
Did I miss anything? Please let me know in the comment box and let’s grow together. Data is not going any where. Let’s make data driven business decisions and not emotional driven business decisions.
#dataanalyst #dataAnalytics #datascience #dataengineer #businessanalyst #databaseadministrators #tableau #sql #powerbi #excel #dataanalystforbeginer
