How I Learned SQL in 4 Months Coming from a Non-Technical Background

How I’ve been learning SQL & developing results as a Data Analyst coming from a business, non-technical background

After transitioning from a Business Development Specialist to a Data Analyst at Upaya, I had lots of catching up to do in terms of gaining technical skills.

Before this, my role was developing business cases and aligning business requirements across different stakeholders along with providing strategy recommendations to the commercial department.

Yet, I also had experience with data analysis but only with Excel and some data visualization tools such as Looker (formerly Google Data Studio).

But, I really wanted to hone in the data analytics domain though it meant I had to start learning many skills from absolute scratch.

My background is in business administration while my work has mostly centered around research & analysis through consulting based projects.

Therefore, building up from these experiences to the current need of the organization in data analytics made sense for me as well as the company.

How I learned SQL

It has been many months now since I first got to know what even the full form of SQL was.

Yet, it was in the 4 month window of time where my main focus was on SQL where I gained this skill.

The 1st step I took was going through the book called Business Analytics for Managers by Gert Laursen & Jesper Thorlund.

It provided me a macro view of the world of business analytics and perspective on where learning SQL & even data analytics as a whole fits into the organization.

The book also clarified to me what the end outcomes could be for the organization by pushing the wheels of data analytics forward.

Yet, these forces helped me most in learning SQL:

1. Doing a Real Project

After reading that book, I took 2 short free crash courses online:

• one from Datacamp called Intermediate SQL Queries
• the other from Udacity called SQL for Data Analysis

It provided me a basis of the technical elements of SQL such as how the queries actually work.

The courses are very interactive & exercise based, so they were fruitful for me.

Especially the course from Udacity provided a strong foundation on the queries used in SQL.

Also, W3 Schools provided clarity when I needed to know about specific new queries, fast & concisely.

Yet, there was an entire project to establish data analytics at Upaya, where SQL queries was the main sauce.

This I would say was the most important part to accelerate learning SQL.

It allowed me to immediately focus on the actual needs of different departments within the company & its stakeholders.

This brings me to my first point which is starting with Intermediate SQL rather than Basic SQL.

I’ve felt that actual use of SQL queries is from intermediate to advanced level in actuality for organizations.

Also, if you start on the intermediate level itself, though at first it might be a bit challenging, quickly you will catch up with all the beginner level stuff too.

Also, a big difference I’ve felt when we’re just learning a skill by ourselves and having external accountability is that naturally the focus goes towards execution in the latter one.

Companies don’t pay you for learning but rather providing new value, so the pressure is always on!

Thus, since there were clear milestones where the output of the data analytics project needed to come within a certain time, 90% plus of the time went towards execution & practice.

2. Focus towards the most important queries

As in anything in life, the 80/20 rule applies to SQL queries too.

20% of the queries are used in 80% plus of the SQL queries.

One benefit of being engaged in a real project with real stakeholders & deadlines that had to be met was that this identification of the most important queries happened more naturally.

If I was doing a project by myself, I would have spent a lot more time than needed on many queries which wouldn’t be used much at all.

From my experience, I would say that you should give unproportionate importance to these queries

• JOINS
• UNION & UNION ALL
• Sub-Queries
• CASE WHEN
• IF
• Aggregations using GROUP BY
• ROW_NUMBER()
• LAG/LEAD

At least for establishing foundation based data analytics in a company, these queries were of high importance in the actual outcomes provided.

Therefore, try and do 10 exercises for each of these queries. It would fasten SQL learnings by a mile.

3. Immediate Feedback

Another benefit of doing a project with real stakeholders is that there is immediate feedback.

There was immediate feedback from the tech team who had created the data warehouse on the accuracy of the data created through queries as well as feedback from the stakeholders.

This helped to clarify what to keep on doing and what to discard.

Especially working with the tech team ensured that the queries were right on track.

Since they too had worked on queries in the creation of the data warehouse & the overall system, the right way to create specific queries was clear and concrete.

I had underestimated the importance of fast feedback before but if you want to bring results effectively, you need to hear the hard truths when you’re in the wrong direction.

There might be rework in the short-term but it’s well-worth it!

Validating your skill level

In order to validate how you’re doing skills wise, one tactic I had used was in applying to jobs and doing interviews, but only top companies.

Especially the technical interview phase in top companies provides a good estimate on how much knowledge you’ve accumulated.

At the same time, it also makes you clear on what areas you are lagging behind in.

This is because top companies carefully prepare specific questions that align with what the data analyst will be working upon.

After 4 months of learning SQL & then doing an interview, I was happy that I had 80% plus proficiency in the technical interview for a tech-giant in Nepal.

I still do not have proficiency in more of the advanced/not very commonly used queries at least so far in my SQL journey such as doing some Window functions with full flow.

I got to know about this while doing technical interviews too.

Summarizing what I did differently

• 80/20 selection
• Starting with real project at intermediate level first, catching up on all theories gradually(reverse order)
• High Stakes Deadlines

Though I was just doing these 3 ideas because of the situation I was presented with, I’ve come to realize that these 3 ideas are very important for accelerated learning.

I realized this while going through Tim Ferriss’ Meta Learning chapter in the book 4 Hour Chef.

He has provided amazing detail on why these 3 ideas work, along with many other ideas and with definite steps.

Results so far

The results so far have been the creation of dashboards by departments & sub-departments at the granular level across the company along with ability to analyze a big part of the data warehouse.

The impact is gradually being seen through data as a key part in the decision making process.

Lots more of skills to add to solidify this though, probably some Python.

Area of improvement I should have included

The one major area where my current process could see improvement would be the availability of a mentor who has gone through my same process.

I would highly recommend this as I believe it would have narrowed down my focus on the most important parts and filter the trivial many even more.

Future plans

Since I’ve tried to fast track my learning process, its important to have repitition in these aspects to really make the learnings concrete.

My approach has been to work in a way where the results are produced on time and fast be it at the expense of distorted complete comprehension.

That’s an area I’ll be working on in the coming days.

Where are you in your data analytics journey?

Let me know by connecting with me on Linkedin here ;)