From data analysts to data science is a viable way to break into the field of data science, and this article aims to explain how to make the transition.
Why become a data analyst in the first place?
I often recommend becoming a data analyst first and then transitioning to a data scientist.
Now, considering that I never worked as a data analyst, why do you do this? OK, for the following reasons.
- Becoming a data analyst is easier than becoming a data scientist.
- You will really learn and understand what business impact data can have – Beginner data scientists often focus on building beautiful models rather than solving business problems.
- In some companies, you can even do the same job as a data scientist, despite the different title differences.
- Time of beat time. So, in my opinion, it’s always better to get into the industry.
A comprehensive roadmap for becoming a data analyst is beyond the scope of this article, but I would love to create an article if you are interested.
What is the difference between a data analyst and a scientist?
Even though data analysts and scientists may be similar in some companies, their roles do vary in most cases.
Typically, data analysts are more subject to business decisions and will use the following tools:
Data scientists will be able to do their best and will have more advanced abilities:
You can think of it as a data analyst who cares more about seeing what is going on, and data scientists who care more about what will happen, such as predicting the future.
You don’t have to transition from data analytics to data science; I know many people are great analysts and are happy in their current roles, getting a lot of sense of accomplishment and being well compensated.
But, I also know that many people want to move to data science and use the data analyst position as stepping stones.
Right or wrong; it depends on your goal. If you are reading this, you want to start jumping, so let’s explain why being a data analyst in the first place is simply not a bad thing.
Develop transitional skills
To go from a data analyst to a data scientist, you need to learn the following.
math
If you are a data analyst, you may already have decent statistical skills, so the main areas you need to focus on are linear algebra and calculus.
- Differentiation and derivatives of standard functions.
- Partial derivatives and multivariate calculations.
- Chain and product rules.
- Matrix and its operations, including trajectories, decision characters, and transpose functions.
coding
As a data analyst, your SQL skills may already be great, so the main features you need to improve are Python and software engineering in general.
- Advanced Python concepts such as unit testing, class and object-oriented programming.
- Data structures and algorithms and system design.
- Understanding of cloud systems such as AWS, Azure or GCP.
- ML libraries such as Scikit-Learn, Xgboost, Tensorflow, and Pytorch.
Machine Learning
You don’t need to be an ML expert, but you should have a good understanding of the basics.
How to learn?
Self-study
The most direct and intuitive way is to study in your spare time, after get off work or on the weekend.
Some people may not like that, but if you want to change your career, it takes time and effort. That is a cruel fact. Many people want to be data scientists, so there is no walking in the park.
There are many resources to learn about the above topics and I have written several blog posts on the exact books and courses you should use.
I’ll link them below and I highly recommend checking it out!
The advantages of self-study are:
- Very cost-effective and even completely free.
- Study according to your own schedule.
- Custom learning paths.
And cons:
- There is no clear structure, so it is easy to make mistakes.
- No formal certificate.
- Need a high degree of discipline and motivation.
Bachelor of Science
You can always return to school and pursue a formal degree in data science or machine learning.
The advantages of this method are:
- Emphasizes mathematics, statistics, computer science and algorithm understanding.
- Some employers have a greater weight in their degrees (especially from top universities).
- Visit teachers, alum network, research projects and internships.
The disadvantages are:
- It may be too heavy to lack real-world projects and data.
- Requires 2-4 years (Bachelor’s degree) or 1-2 years (Master’s degree).
- May be expensive
- A strong academic record is required, which may be a GRE, a letter of recommendation or a prerequisite course.
Training Camp
These have been everywhere in recent years due to the growing demand for data and machine learning roles.
Overall, they offer cheaper degree alternatives and offer more hands-on programs and hands-on courses.
Professionals are:
- Most boot camps are 3-6 months long and focus only on data science skills.
- A lot of attention is paid to real-world projects, coding and tools (Python, SQL, machine learning libraries).
- Many offer career coaching, resume reviews, mock interviews and job placement support.
- Cheaper than a degree.
And cons:
- The theoretical depth is shallow.
- It’s paced too fast.
- The quality may vary, so be sure to do your research before attending.
- Limited reputation for employers.
In your current job
This is my favorite, and it is the most effective and worthwhile.
If you work on the right project and you can also express your interest in the skills and tools to develop to the manager, you can learn everything in your current job.
When their direct reports take initiative and show enthusiasm for their work, managers like it because it also benefits them as a byproduct.
Professionals are:
- Get paid to study, win!
- Access to actual data and business issues.
- Real-life data science experience can be added to your portfolio.
- It even allows you to transition to data science full-time.
The disadvantages are:
- This may result in more workloads.
- Role expectations may be fixed and may have little internal liquidity.
Create your portfolio
During and after your study, you need to create some evidence for data scientists about what you can do, which is basically making a portfolio.
I plan to release a more in-depth video soon on what a strong data science portfolio should include. But for now, here’s the short version:
- Kaggle Competition– Make one or two. This is not to put high; it is to show that you can use the real dataset and follow.
- 4–5 simple projects– These should be quick builds that you can do in one or two days. Upload them to github. Better yet, write a short blog post to explain your process and decisions.
- Blog Posts– Aim at about five. They can cover anything related to data science: tutorials, insights, lessons learned – just show that you are thinking and communicating well in critical thinking.
- A solid personal project– This is your core. You work more deeply for one month, one or two hours a day. It should show end-to-end thinking and be something you are really interested in.
That’s it.
People beat this step too many. Just start building – keep coming.
Find a job
As I said above, the easiest way is to transition internally.
If this is not an option, then you need to be busy applying!
You need to align your resume/resume, LinkedIn profile, and GitHub account with your job role as a data scientist. Make sure you start calling yourself a data scientist, rather than “ambition.”
I studied physics in college but never received any compensation for practicing physics. I’m still a physicist. The same is true for data science.
You can use your portfolio anywhere to demonstrate your abilities. Your GitHub profile should be linked to your LinkedIn profile and then to your blog posts and other related content. Get an ecosystem that induces people so that they can “spend more time” with you.
After everything is fully prepared, start applying for more analytics-centric roles with the title data scientist. Of course, you can choose more machine learning methods, but they are hard to get.
Also use your network for recommendations. If you have been working in a data field for a while, you must have at least one person who knows you can recommend you to a data science job.
The beauty of transitioning from a data analyst to a data scientist is that you can take a moment because you are already making money and making money on the spot, which relieves the stress. Just make sure you stick with it and make consistent progress!
Another thing!
I offer a 1:1 coaching call where we can chat with anything you need – whether it’s a project, career advice or just figuring out your next step. I’m here to help you move forward!
1:1 Guidance phone number with Egor Howell
Career guidance, job advice, project assistance, restoration reviewtopmate.io
