Ian Klosowicz
You don't need a degree to become a data analyst. You need SQL, a couple of projects, and a job search strategy that gets your resume in front of humans instead of dying in an ATS.
I know this because it's exactly what I did. No relevant degree. No bootcamp. I taught myself on nights and weekends, built a few projects, and got hired as a data analyst about a year after I decided to make the switch. I now work in data engineering and have watched thousands of people try to follow a similar path — some successfully, most making the same avoidable mistakes.
This is the actual step-by-step of how to do it.
SQL is the single most important skill for a data analyst role. Every other tool is secondary to this one. If you only have time to learn one thing before you start applying, it's SQL.
The level you need isn't advanced. You need to be able to:
That covers 80 to 90% of what a junior analyst does in SQL on a daily basis. Window functions, CTEs, and subqueries are useful to know and will come up, but they're learnable on the job once you have the fundamentals solid.
The mistake most people make is treating SQL like a course to complete rather than a skill to practice. Finish the fundamentals fast, then spend most of your time writing actual queries against real datasets. That's what builds fluency.
Free resources that work: Mode's SQL Tutorial, SQLZoo, and the practice sections in Khan Academy's SQL course. Pick one and stick with it. Switching between 4 courses is procrastination with extra steps.
Excel is listed on nearly every analyst job description. When companies say "proficient in Excel," they don't mean you've memorized 400 formulas. They mean you can do useful things with data quickly.
The parts that actually matter:
You don't need to learn everything. You need to learn the parts analysts actually use. That's a much shorter list than most Excel courses suggest.
Google Sheets works fine if you don't have an Excel license. The core functions are almost identical. Knowing either one is enough for most roles.
Tableau and Power BI are the two most common BI tools in job postings. You don't need to know both. Pick one and build something real with it.
Tableau Public is free and is the most commonly used starting point because you can publish your work directly to a public URL. Power BI Desktop is also free. Either works.
The goal at this stage isn't mastery. It's being able to connect a dataset, build a few charts, and assemble them into a dashboard that tells a coherent story. That's what hiring managers are looking for when they see a BI tool on your resume — evidence that you can build something, not proof that you've watched 40 hours of tutorials.
Most BI tools are taught on the job anyway. Companies have their own setups, their own data sources, their own templates. What they're hiring for is someone who can learn a new tool quickly. Showing up with one real dashboard is proof you can do that.
This is the most important step and the one most people delay longest.
3 projects is the right number. 1 is not enough to show range. 5 is too many to maintain well. 3 lets you cover different datasets, different tools, and different business questions without spreading too thin.
What makes a good project:
The written summary is the part most people skip. It's also the part that separates candidates who can do analysis from candidates who can communicate analysis. Hiring managers are looking for both.
I got hired with 3 projects on my resume — 2 Tableau dashboards on Tableau Public and 1 Power BI project, each at its own link. No portfolio site tying them together. Just 3 links on a resume and the work spoke for itself.
Good project areas if you're not sure what to build: e-commerce sales analysis, customer churn analysis, marketing funnel analysis, or any dataset from an industry you have prior experience in. Domain knowledge from a previous career makes your analysis more credible, not less.
If you want project prompts, templates, and step-by-step guidance for building all 3, Analyst Hive covers exactly that as part of the first month of the program.
Recruiters source from LinkedIn constantly. A weak profile means you're invisible to the people actively looking to fill roles.
The things that matter most on LinkedIn for a career changer without a degree:
LinkedIn also matters for networking. Connecting with data analysts at companies you want to work at, reaching out to recruiters who post analyst roles, and commenting on posts from people in the field all build the kind of visibility that gets you in conversations before a job is even posted.
Without a relevant degree, your resume has to lead with what you can do. The structure that works:
Keep it to one page. Hiring managers spend 6 seconds on a first pass. The skills and project links need to land in that window.
One thing that genuinely helps: tailor the skills section slightly for each role. If the job posting mentions Python, and you have basic Python exposure, include it. If it mentions a specific BI tool you know, name it. ATS systems keyword-match, and small adjustments make a real difference in whether you clear the automated screen.
The job search is where most self-taught analysts stall. They build the skills, make the projects, update the resume, and then apply to 10 jobs and wait. That's not a job search. That's a lottery ticket.
A real job search looks like this:
I went through a stretch of rejections before the right opportunity came through. What broke it was one listing, one conversation, and a few interviews that went well. The job search is not a reflection of your ability. It's a numbers game with a skill floor. Clear the floor, apply at volume, and the right role comes through eventually.
If you want a structured day-by-day plan for all of this — the skills, the projects, the LinkedIn setup, the resume, and the job search — the Analyst Hive program is built exactly for this situation. It's what I wish had existed when I was making this transition.
Realistically, 3 to 6 months of consistent daily work to get job-ready. Getting the actual offer adds time on top of that depending on how competitive your market is and how hard you're running the job search.
I took about a year from deciding to pursue data to getting hired. I wasn't doing it full-time — I was building skills and applying while working another job. People who treat it more like a full-time effort tend to move faster.
The variables that affect timeline:
There's no shortcut that changes the fundamentals. The skills have to be real, the projects have to be real, and the job search has to be run at real volume. Everything else is details.
What's the first thing I should learn to become a data analyst?
SQL. Start there before anything else. It's the skill that appears in almost every analyst job description, it's the one most likely to be tested in a technical screen, and it's the foundation that makes everything else — BI tools, data cleaning, project work — easier to pick up. Spend 4 to 6 weeks getting solid on the fundamentals before moving on.
Do I need to learn Python to get a data analyst job?
For most entry-level analyst roles, no. SQL and Excel cover the majority of what junior analysts do. Python comes up more in analyst roles that lean toward data science, and those postings usually say so. If Python isn't listed in the job description, don't block your job search on it. Learn it after you're hired if the role calls for it.
How many projects do I need in my portfolio?
3 is the right number for most people. It's enough to show range across different datasets and tools without being so many that maintaining them becomes a job in itself. Each project should have a live link you can put directly on your resume. A portfolio site is not required — individual project links are enough.
Should I get a certification like the Google Data Analytics Certificate?
A certification alone doesn't move the needle much with hiring managers. What matters is what you produce during and after it. If a certification gives you structure and helps you build real projects, it's worth doing. If you're treating the certificate as a credential that substitutes for project work, it won't do what you're hoping. Lead with the projects. Stack the cert underneath.
How do I get experience if no one will hire me without experience?
Build it yourself through portfolio projects. Use public datasets to answer real business questions, publish the work, and link it on your resume. Projects are entry-level experience you create without needing an employer. The technical screen at most companies doesn't care whether your SQL came from a job or a personal project — it tests the same skill either way.
