Three years ago my cousin called me in a panic. She just got laid off from her retail management job and someone told her just become a data analyst they make good money and companies are hiring like crazy. She had zero coding experience a business degree from 2015 and about four months of savings to figure this out.

I was working as a data analyst for a mid size healthcare company at that point so she asked me to walk her through it. What I told her that night turned into months of back and forth texts a few zoom screen shares where I watched her struggle with her first pivot table and eventually about seven months later an offer letter from a logistics company in Ohio.

This article is basically that conversation cleaned up a bit and written down.

Why Everyone Suddenly Wants This Job

Data analyst roles blew up for a real reason. Companies collected way more data than they knew what to do with. Sales numbers website clicks customer complaints inventory counts. Somebody needed to make sense of all of it.

The appeal is obvious once you look at it from a career switcher’s view. You don’t need a computer science degree. You don’t need to be some math genius. You just need to be curious decent with numbers and willing to learn a few tools.

But here’s what nobody tells you upfront. The market got flooded with people trying to switch into this field around 2021 to 2023 and now hiring managers are way more picky. You can see it just by scrolling job postings on LinkedIn Jobs or Indeed where a single analyst posting can pull hundreds of applicants in a couple days. My cousin got rejected from over 40 applications before she even got one interview. That’s normal now not a sign something’s wrong with you.

What a Data Analyst Actually Does (Not the Textbook Version)

Forget the LinkedIn definition for a sec. On a typical Tuesday here’s what my job actually looks like.

Someone from marketing messages me asking why email open rates dropped last month. I pull data from our CRM cross reference it with a spreadsheet marketing maintains notice a chunk of it doesn’t match up spend 40 minutes figuring out why (turns out someone changed a naming convention) fix it then build a chart showing the actual trend.

That’s the job basically. It’s like 60% detective work 30% cleaning messy data and maybe 10% actual “analysis” that looks impressive in a slide.

If you’re picturing yourself building fancy machine learning models day one just adjust that expectation now. Most entry level and even mid level jobs are about answering business questions with data that’s usually incomplete poorly organized or scattered across three different systems. If you wanna see what employers actually expect day to day the Bureau of Labor Statistics has a decent breakdown in its Occupational Outlook Handbook.

The Skills That Actually Matter

1. Excel still unironically

Sounds unglamorous I know but Excel or Google Sheets is where most companies actually live. VLOOKUP INDEX MATCH pivot tables basic formulas. If you can’t do these fast you’re gonna struggle even after learning the fancier stuff.

2. SQL

This one’s the real gatekeeper. Almost every analyst posting asks for SQL cause most business data sits in databases. You need to know how to write queries to pull filter join and group data. Honestly SQL got me more interviews than anything else on my resume.

Free places to practice, Mode Analytics’ SQL tutorial and a site called SQLZoo. My cousin used LeetCode’s SQL section too though that’s more for tech company style interviews. StrataScratch is another good one with more realistic business questions.

3. A visualization tool, Tableau or Power BI

Pick one don’t try learning both at once. Power BI is more common if you’re going for companies that use Microsoft stuff heavily which is a lot of them. Tableau shows up more in tech and consulting.

4. Basic Python, optional but helps

You don’t need to be a programmer or anything. Just enough Python to clean data with pandas maybe automate a boring report. I use it like twice a month honestly. It’s more of a nice to have that makes you stand out than something you use daily. Kaggle’s free Python course is a fine place to start if you need structure.

5. Communication, seriously underrated

The best analysts I’ve worked with weren’t the ones with the fanciest technical skills. They were the ones who could explain here’s what the data means and here’s what you should do about it to someone who doesn’t care about the technical side at all. If you can’t turn numbers into a story you’ll plateau fast.

What the Money Actually Looks Like

Salary depends a lot on city industry and experience so take these as rough numbers based on what I’ve seen among people I know and postings I’ve looked through.

• Entry level (0 to 2 years), roughly $55,000 to $70,000
• Mid level (2 to 5 years), around $70,000 to $95,000
• Senior analyst, $95,000 to $130,000 or more especially in tech finance or healthcare

Location matters a ton. A junior analyst in San Francisco or New York might start around $75K while the same job in a smaller Midwest city might be $52K. Remote roles complicate this too, some companies pay based on your location others pay one flat rate no matter where you live.

Industry makes a difference too. Finance and tech tend to pay more than retail or nonprofit for the same kind of work.

I’d recommend checking current numbers on Glassdoor or Levels.fyi for tech specific roles before negotiating anything since these numbers shift every year. Payscale is another decent one to cross check against.

The Step By Step Path That Actually Worked

Here’s basically what my cousin did in order.

Step 1: Learn Excel properly first (2 to 3 weeks) Not just basic formulas, pivot tables VLOOKUP conditional formatting charts. She used free youtube tutorials from a channel called Leila Gharani.

Step 2: SQL basics (4 to 6 weeks) Used the free Mode SQL tutorial then practiced on StrataScratch which has real interview style questions.

Step 3: Pick one visualization tool (3 to 4 weeks) She went with Power BI since more local postings mentioned it over Tableau. Microsoft has a free learning path for it.

Step 4: Build 2 or 3 portfolio projects This is where most people either shine or get stuck honestly. She used public datasets, one from Kaggle about retail sales another using data from a local city’s open data portal about parking violations. Real ugly unglamorous data. Not another Titanic dataset project every recruiter has already seen 500 times.

Step 5: Put it on GitHub with a simple write up Doesn’t need to be fancy just make a free account on GitHub and write what question you were answering what you found and what you’d recommend. A README explaining your thinking matters way more than perfect code.

Step 6: Tailor the resume to each job She stopped sending the same resume everywhere and started matching keywords from each posting. Tools like Jobscan can help check how well your resume lines up with a job description. This alone seemed to double her interview rate.

Step 7: Apply consistently don’t burn out She set a goal of 5 applications a day not 50 in one day then nothing for a week straight. Consistency mattered way more than doing everything in one burst.

Mistakes I Watched Her Make

She spent almost three weeks trying to learn Python deep before even touching SQL because some youtube video ranked it as the best language for data analysts. Total waste of time for her situation. Most jobs she applied to barely mentioned Python but heavily wanted SQL and Excel.

She also built her first portfolio project with a super clean pre processed dataset. Looked nice but taught her nothing about real world messiness. Her second project using genuinely messy data pulled from data.gov with missing values and inconsistent formatting actually came up in her interview cause the interviewer asked walk me through a time you dealt with messy data and she had a real story ready.

Another mistake, not customizing her resume. First month she used one generic resume everywhere. Once she started tailoring it even just swapping a few keywords and reordering skills based on the posting her callback rate noticeably went up.

Last thing, she almost gave up after around 30 rejections with zero interviews. That’s the point where a lot of people quit honestly. She pushed through one more round after tweaking her resume format, switched from a fancy canva template to a plain ATS friendly word doc, and interview requests started coming within two weeks.

Where People Actually Find These Jobs

Beyond the obvious (LinkedIn Indeed) a few things worked better than expected.

Company career pages directly. Sometimes jobs get flooded on LinkedIn but the same posting on the company’s own site has way less competition.

Local networking events or meetups even virtual ones. Sites like Meetup still have active data groups in most bigger cities. She got one interview cause someone she met at a virtual data meetup forwarded her resume internally.

Referrals matter more than people admit honestly. If you know anyone even loosely at a company you’re targeting ask them to refer you. Employee referrals often get looked at before random applications even get opened.

Final Thought

If you’re thinking about switching into this it’s genuinely doable without some fancy degree but it’s not a quick three week bootcamp fix either. Took my cousin about seven months of consistent sometimes frustrating effort to get there.

The people who actually make it through aren’t necessarily the smartest or most technical. They’re the ones who kept applying after rejection number 25 who built real projects instead of just watching tutorial after tutorial and who got comfortable explaining numbers to people who don’t care about the numbers themselves just what they actually mean.