One writes SQL at 2 AM. The other rewrites the requirements doc at 2 AM. Both make $90K+. Here is how to pick.
Get new posts delivered to your inbox. No spam, unsubscribe anytime.
I've held both titles. Data Analyst at Unibank, then a role that blurred into business analysis, then back to pure data work. The job postings looked almost identical. The actual work couldn't have been more different.
At one desk, I spent my mornings writing SQL queries to extract loan performance data, my afternoons building dashboards, and my evenings debugging a Python ETL script that broke because someone changed a column name upstream. At the other desk, I spent my mornings in stakeholder meetings, my afternoons writing requirements documents, and my evenings rewriting those documents because the VP changed their mind.
Same company. Same floor. Completely different jobs. And yet, half the articles comparing these roles make them sound interchangeable. They're not.
Let's start with money because that's what you actually Googled.
Data Analysts earn a median total pay of $92,000 according to Glassdoor, with the typical range falling between $72,000 (25th percentile) and $122,000 (75th percentile). The average jumped to $111,000 in Q1 2025 — a $20,000 increase from early 2024.
Business Analysts earn a median total pay of $105,000 according to Glassdoor, with a range between $84,000 and $137,000. Senior business analysts in tech or finance command $138,000+, with top earners in major metros exceeding $150,000.
Business analysts earn about 15% more at the median. But here's the part nobody mentions: the gap narrows fast when data analysts pick up ML or AI skills. A data analyst who can build a predictive model is suddenly competing for $130K-$160K roles. A business analyst who can't write SQL is stuck at the $105K ceiling.
The real question isn't which title pays more today. It's which ceiling you're willing to live under.
Data science careers are projected to grow at 36% from 2023 to 2033. Management analysts (the BLS category that includes business analysts) are growing at 11% over the same period.
That's a 3x difference in growth rate.
The global data analytics market is projected to hit $104 billion by end of 2026, growing at 21.5% annually. Industry forecasts suggest nearly 11.5 million new jobs in data science and analytics by late 2026.
Does that mean business analysis is dying? No. It means the floor is rising for data roles while business analyst demand grows at a slower, steadier pace. Both are safe bets. One is a faster escalator.
Search "data analyst vs business analyst" and you'll find a hundred articles that give you a clean two-column table: data analysts work with data, business analysts work with stakeholders. Technical vs. non-technical. Python vs. PowerPoint.
That's a 2018 understanding of 2026 roles.
Here's what's actually happening: the roles are converging from opposite directions. Data analysts are being asked to present findings to executives and tie insights to business outcomes. Business analysts are being asked to write SQL queries and build their own dashboards instead of requesting them from the data team.
The pure "I just write SQL" data analyst is becoming rare. The pure "I just write requirements" business analyst is becoming rarer. The market wants hybrids.
Stakeholder communication now appears in nearly 60% of data analyst job postings. That's not a data skill — that's a business analyst skill bleeding into a data role. Meanwhile, SQL appears in roughly 85% of data analyst interviews and increasingly in business analyst postings too.
The most valuable people in either role are the ones who can do both: pull the data AND explain what it means to someone who doesn't care about your JOIN syntax.
This is where the difference becomes visceral. Forget the job description — here's what Tuesday looks like.
9:00 AM — Check your Slack. The marketing team wants to know why conversion rates dropped 12% last week. You open your SQL editor.
9:30 AM — Write a query joining the events table with the campaigns table, filtering by date range. Find that the drop correlates with a new landing page deployment. You dig deeper.
10:30 AM — Build a quick visualization in Tableau showing the conversion funnel before and after the landing page change. The breakpoint is clear: users drop off at the pricing section.
11:00 AM — Standup meeting. You present your finding in 90 seconds. The PM asks for a dashboard they can monitor.
1:00 PM — Spend the afternoon building that dashboard. You're wrangling a messy dataset — duplicate entries, null values, inconsistent date formats. You write a Python script to clean it.
3:30 PM — The ETL pipeline broke. A source table schema changed overnight. You fix the pipeline, add a validation check so it doesn't silently fail again.
5:00 PM — Push your dashboard to production. Write documentation.
Your tools today: SQL, Python, Tableau, a terminal, and Stack Overflow.
9:00 AM — Requirements meeting with the product team. They want to redesign the checkout flow. You're taking notes, asking clarifying questions, identifying edge cases nobody thought about.
10:30 AM — You document the requirements in Jira. User stories, acceptance criteria, process flow diagrams. You map out the current checkout flow vs. the proposed one.
11:30 AM — Meeting with the engineering team. You walk them through the requirements. They push back on three items — one is technically infeasible, one needs more detail, one conflicts with an existing system. You negotiate.
1:00 PM — Update the BRD (Business Requirements Document) based on the morning's feedback. Add wireframe annotations.
2:30 PM — Pull some data from the analytics dashboard to validate the hypothesis that checkout abandonment is happening at the payment step. You don't write the query — you ask the data team, or you use a BI tool with a pre-built view.
3:30 PM — Stakeholder alignment meeting. You present the finalized requirements to the VP. She wants one more change. You update the docs.
5:00 PM — Send the updated BRD to all stakeholders. Prep for tomorrow's sprint planning.
Your tools today: Jira, Confluence, Excel, Miro, PowerPoint, and your inbox.
Data analysts answer: "What happened and why?" Business analysts answer: "What should we build and how should it work?"
One spends most of their time with data. The other spends most of their time with people.
| Skill | Data Analyst | Business Analyst |
|---|---|---|
| SQL | Required (daily use) | Helpful (growing requirement) |
| Python/R | Expected | Rarely needed |
| Statistics | Core skill | Basic understanding |
| Tableau/Power BI | Primary output | Consumption, not creation |
| Excel | Heavy use | Heavy use |
| Jira/Confluence | Occasional | Daily |
| Requirements writing | Never | Core skill |
| Stakeholder management | Growing need | Primary skill |
| Process modeling | Rare | Core skill (BPMN, UML) |
| Wireframing | Never | Often |
| Presentation skills | Important | Essential |
| Domain knowledge | Important | Essential |
The education paths differ too. Data analyst roles increasingly expect technical credentials — statistics, computer science, or a data analytics degree. Business analyst roles favor business administration, management, or an MBA, though many enter from any background.
Here's the interesting stat: SQL appears in 85% of data analyst interviews. For business analysts, the number is closer to 30-40%. But that gap is closing fast.
The interviews test completely different muscles.
You'll get:
The emphasis is on can you find the answer in the data?
You'll get:
The emphasis is on can you get the right thing built?
One tests your relationship with data. The other tests your relationship with people.
The paths diverge more than people think.
Data Analyst path: Junior Data Analyst → Data Analyst → Senior Data Analyst → Lead/Principal Analyst → Analytics Manager
Or the technical branch: Senior Data Analyst → Data Scientist → ML Engineer → Staff/Principal ML Engineer
Business Analyst path: Junior BA → Business Analyst → Senior BA → Lead BA → Business Intelligence Manager
Or the leadership branch: Senior BA → Product Manager → Director of Product → VP of Product
Here's the key insight: data analysts have a technical escape hatch. If you build strong Python and ML skills, you can pivot to data science or ML engineering — roles that pay $150K-$250K+. The technical ceiling is much higher.
Business analysts have a leadership escape hatch. The BA-to-PM pipeline is well-established. Product managers at top companies earn $150K-$200K+, and the path to Director/VP is clearer from a PM role than from a data role.
Pick based on where you want to end up, not where you start.
Choose Data Analyst if:
Choose Business Analyst if:
Choose neither if:
I said earlier that the roles are converging. Let me be specific about why this matters for your career.
The highest-paid professionals in both fields are hybrids. A data analyst who can present to executives and tie findings to revenue impact earns more than one who just writes queries. A business analyst who can pull their own data and validate assumptions with SQL earns more than one who needs to request every data point.
The market calls these people different things — "analytics engineer," "product analyst," "business intelligence analyst" — but the skill profile is the same:
If you're starting out, I'd invest in SQL and one BI tool regardless of which title you pursue. Those skills are table stakes for both roles and will be for the foreseeable future.
I've been on both sides. Here's my honest take.
Business analysts are overpaid relative to their technical depth. There, I said it. A median of $105K for a role that primarily involves writing requirements documents and attending meetings feels inflated — especially when a data analyst doing more technical work earns $92K. The premium exists because BAs sit closer to decision-makers and their work is more visible to leadership. Visibility pays.
Data analysts are underpaid relative to their future value. A data analyst with strong SQL, Python, and statistics skills is 6-12 months of focused learning away from a data scientist role paying $130K-$160K. The technical foundation compounds. A BA with no coding skills has a harder time making that jump.
The BA role is shrinking in some sectors. In tech companies, the traditional BA role is being absorbed by product managers and data analysts. The standalone "business analyst" title is more common in banking, consulting, and healthcare — industries with complex regulatory requirements that demand dedicated requirements documentation. In a startup? There is no BA. The PM and the data analyst split the work.
The DA role is expanding. Every company is becoming a data company. The 36% growth projection isn't hype — it's driven by real demand from healthcare, fintech, retail, and every industry that generates data (which is all of them).
My actual recommendation: if you're technical and enjoy coding, go data analyst. The ceiling is higher, the growth is faster, and you keep optionality for data science and ML roles. If you're a people person who thinks in process flows and enjoys the politics of getting organizations to agree on things, go business analyst. The PM pipeline is your real exit strategy.
Either way, learn SQL. That's not optional anymore for either role. It shows up in 85% of DA interviews and increasingly in BA interviews too.
The worst career move is picking based on the title and ignoring what you'll actually do from 9 to 5. I've seen miserable data analysts who should have been BAs, and frustrated BAs who should have been writing Python. Pick the work, not the label.