Ever wonder how Spotify chooses the perfect song every time? Or how Netflix finds just the right movie for a lazy Sunday? This isn’t a creepy coincidence – merely the work of data analysts. Nearly every company nowadays uses data analysis, making it an extremely in-demand career.
Breaking into this field can seem difficult — but we’ve got you covered. In this career guide, you will learn what data analysts do, how much they earn, required skills, and how to master them. The guide is intended to help you avoid any pitfalls in taking your first steps toward a career in data analysis.
What is data analysis?
In short, data analysis is about turning raw data into insights that help a company understand what happened, why, and what to do about it. In other words, it is a process of finding, collecting, validating, analyzing, and interpreting data to help a company make more informed decisions. This method applies, regardless of industry: every company uses data and wants to make better decisions.
What does a data analyst do?
Based on the definition above, a data analyst:
- Identifies the true needs of the requestor
- Understands what sets of data to analyze
- Retrieves relevant info from the databases
- Cleans and validates the data
- Analyzes the data (Hence the name!)
- Interprets the results
- Visualizes the outcomes in a clear, non-technical, and engaging way
- Proposes a solution
Which age group should be targeted during our next ad campaign? What regions are most vulnerable to forest fires? What medicine should be stocked to prevent the seasonal spread of a disease? These are the types of questions data analysts answer.
Let's look at some examples that illustrate the process in more detail.
A data analyst works for a company that rents out housing through Airbnb
The company rents out housing for short-term sublease through Airbnb. First, they search for property owners to conclude long-term lease contracts. Then, they sublease the properties to tenants.
Currently, the company uses the Airbnb website itself for analysis, and manually transfers data to Google Sheets. They would like to do something to facilitate and speed up the search for profitable long-term rental properties.
The users of the dashboard are the head of the rent department and subordinate managers. It will be used daily, on office desktops.
The data analyst interviews the stakeholders and determines their needs, because the initial request is vague.
After coming up with the list of tasks and validating it with the requestor, the analyst retrieves the data.
Using SQL, he or she gets six files in tabular format with info about dates, listings, listings summary, districts, reviews, review summaries, and also one non-tabular file with geodata.
The analyst then cleans the data (e.g. filters the listings with districts that don’t correspond to actual geodata), validates it with the requestor, and composes the dashboard in Tableau.
The resulting dashboard consists of five blocks:
- “Occupation Rate by Neighborhood” displays:
Average annual occupancy in the best district
Most popular districts
- The “Factoid” block displays:
Average rental cost in the selected district
Average annual occupancy in the selected district
Number of unique listings in the selected area
- The "Best Listings" block to compare and choose the best offers by:
Average user rating
Average annual occupancy
Potential annual revenue from the estate
- The "Top 10 Listings by Potential Annual Revenue" with clickable offers that land on the property’s web page.
- The filtering block allows a requestor to filter data by district, average annual occupancy, and the number of reviews over the past year.
After validating the dashboard with the requestor, the analyst performs a demo session with functionality tests on different PCs.
As a result, the department's staff uses the dashboard on a daily basis. Google Sheets and analyzing the Airbnb website are things of the past.
A data analyst works in a company that makes mobile games
- Create a function that will count the retention of players (by days, from the date of registration of the player).
- Compose a set of proposals according to the results of the last A/B test (An A/B test is a method of comparing two versions of an app against each other to determine which one performs better.)
- Offer metrics to evaluate the results of the last thematic event in the game.
Three CSV files with info about registration time, time when users logged into the game, and A/B test results.
During the research, the analyst discovered several anomalies:
- Irrelevant user registrations dated 1998–2013 (1.76% of the total number of registrations). They were deleted from the research.
- About 7% of the data had an abnormal frequency of visits to the game: less than five minutes between sessions (possibly bots or random user visits to the game, without the game process itself). They were removed from the dataset.
- When analyzing data for A/B testing, 128 abnormally high values of revenue per user in the control group were found — these values were removed from the data, because they are most likely erroneous (technical bug, problems with splitting, etc.)
Results of A/B tests
- The conversion into purchase and the distributions of revenue values by user in the test and control groups are statistically the same.
- The average revenue per user, the average revenue per paying user, and the distribution of revenue values from paying users in the test group is statistically much higher in the test group, compared to the control group.
The analyst comes to three conclusions:
- The maximum retention of users occurs between day four and day seven of the product usage, and after that the indicator begins to fall. New promotions or new game mechanics should be implemented to increase retention after the seventh day.
- A list of metrics for evaluating the results of the past thematic event should be the following: average session duration, daily and monthly active users, stickiness rate, users online, average transaction value. These are optimal for monitoring how changes in game mechanics affect user behavior and product development.
- Based on the A/B test, the option used in the test group should be implemented in a live product.
Data analyst skills
To effectively solve the problems above, a data analyst should master a set of hard and soft skills. Unlike the business analyst role, where soft skills prevail over hard ones, the data analyst domain is more on the tech side. Let’s look at the hard skills first.
Data analyst hard skills:
- Spreadsheets, SQL. Good analysis starts with understanding how to use spreadsheets and SQL databases to retrieve and filter data from these spreadsheets. That said, an analyst should master the spreadsheets: know how to clean and filter data in them, how to use pivot tables, graphs and charts. He also needs to be well-versed with business analytics, e.g. metrics and funnels, cohort analysis, unit economics, and user metrics. When these skills are in place, the analyst can start working in SQL to query data from the spreadsheets, find relationships between tables, work with functions, and compose advanced queries.
- Data visualization. Data analysis is just a set of incomprehensible numbers and values until you visualize it, right? And the more interactive and engaging for the stakeholders, the better. The analyst should know how to build interactive charts in Tableau, choose the right visualization, and work with Tableau dashboards.
- Python. Analysts typically have a solid foundation in Python and software development tools used with Python. Variables, strings, lists, data types, arithmetiс operations, loops, functions, command lines, sophisticated libraries like Pandas and Matplotlib — these are the elements of the skillset required.
Already sounding complicated? We understand. But at Practicum, we’ve got you covered: our community managers, code editors, tutors, senior students, tech support, and career coaches will constantly guide you and help solve any issues.
Over 1K+ Practicum graduates know for sure that even a person with no experience in tech can master these skills and become a data analyst. It is a matter of perseverance, not background.
But if in some (very unlikely) case you don’t get a new job in tech within six months of graduation, we will refund 100% of your Practicum tuition.
Data analyst soft skills:
- Problem-solving and critical thinking. Data analysts need to have a good understanding of the question being asked and the problem to be solved. Based on the info he or she gets from the requestor, an analyst should retrieve a relevant set of data and find patterns or trends that might reveal a problem.
- Presentation. The ability to get ideas across to coworkers, work on a team, and explain your decisions is a must.
- Data curiosity and business area knowledge. Effective data analysts enjoy sifting through data and making hypotheses. Knowing about the business domain helps an analyst to get familiar with the main issues and trends (e.g. a person with a background in banking is likely to be more effective as a data analyst in this sphere.)
Now back to our data analysts working at rent and gaming companies. To solve their problems, they had to understand business processes and make competent conclusions. This was made possible with the use of particular tools. They also used good soft skills to validate their proposals:
- The first data analyst had to initiate a meeting with the head of the rent department, because the initial request was vague: "We work in Google Sheets and directly in Airbnb. It takes a lot of time. Can you do something with it?" As a result of the meeting, the analyst found out the true needs of the customer and created a dashboard in Tableau with five blocks.
- The second data analyst demonstrated data curiosity skills: while working with the data, they came up with a conclusion that it would be helpful to conduct additional research on how the metrics above correlate (or even predict) user spending and lifetime value.
The data analyst’s career path
What makes data analysis great is that everybody needs it. These days both tech and non-tech companies have loads of data, and need those who can interpret it. Insurance, banking, retail, healthcare, gaming, and other companies choose a data-driven approach and employ data analysts to improve proposals and decision-making. With the many opportunities available across industries, you have the leeway to pursue your passion in pretty much any domain.
Practicum’s alumni work in both established tech companies and startups. Here are some of them:
Data analysis career paths depend on skills, state, company, and level of responsibility.
The median salary of a Practicum graduate is $65,500. As they grow in status and experience, they can earn $75,000+.
Every company has different career growth and review processes. Large companies have more structure with formal reviews and expectations, and a ladder with objectives and support for meeting them. Small companies don’t have (as much of) that — so a data analyst can choose to grow, or not, but either way, they have to be a bit more proactive to figure more of it out themselves. This will allow them to maximize their growth in whatever style of company they choose. There is also the option in data analysis careers, depending on the field, to become more specialized. This adds specific tools to general knowledge in data analysis. For example, financial analysts gain new knowledge in finance and investments, and marketing analysts acquire new marketing tools and metrics.
- The data analyst in a renting company began to spend more time studying business processes and their impact on the company's growth. So he became a business analyst.
- The data analyst in a gaming company enjoyed doing research and studying the impact of external circumstances on customers' buying power. So he became a marketing analyst.
How to become a data analyst
So how can Practicum help you break into the exciting field of data analytics? We’ve broken it down into six steps:
1. Decide what tech domain resonates most with you.
Have a researcher's approach, like to surf data, and read between the lines? If yes, you might be a good fit for this role. (Still not sure what tech job works for you? Take our quiz!)
Keep in mind that you don’t need a tech background. All you need is а love for discovery and persistence. Also note that you don’t need to stick to the data analyst’s role forever: career paths in the tech domain are pretty flexible, and analysts often become data scientists, for example.
2. Take Practicum’s top-rated Data Analytics Bootcamp (Named the best coding bootcamp three years in a row!)
The learning process is divided into sprints, or two-to-three-week periods of work towards a tangible outcome. Most tech companies work in sprints, so you'll show up prepared.
Each sprint caps off with a project that will be reviewed line-by-line by a professional software engineer.
All students take on externships — real-life projects at actual companies. At the end of the program, you’ll have a polished, 12-project portfolio.
Practicum prepares "universal soldiers" in data analysis: you’ll have a solid skillset to use in the domain you’re passionate about. Business, banking, healthcare, environment protection, drones, games — you name it.
Aside from the "best bootcamp" status, our program boasts the best value for your money.
3. Сreate a resume, LinkedIn profile, and GitHub account that sells.
In our Career Prep Course, you will learn how to write a standout resume and cover letter, create a professional LinkedIn profile and GitHub portfolio, and rock your job search. Our one-on-one sessions with a career coach will help you develop a job search strategy.
4. Highlight fields that interest you, find the companies that closely match your preferences, and try to get a reference from your future coworker (It’s a great leg up in a job search — 60% of people find jobs in IT through referrals).
5. Prepare for an interview.
Take advantage of your 4+ free hours of mock interviews. These will give you the confidence to ace a real one!
6. Land the job of your dreams, like 85% of our graduates!
And after you do so, we'll stay in touch to make sure you're confident in your new position. Just because we care =)
Schedule a call with an advisor and start your data analytics career today!