The Bureau of Labor Statistics pegs the median data scientist salary at $108,020. That number is both accurate and nearly useless — it flattens entry-level analysts in rural markets with staff ML engineers at Stripe into a single figure. If you're trying to figure out what you'll actually earn, or what it takes to move up a bracket, you need the breakdown.
This guide covers data science salary ranges by experience level, city, specialization, and industry — using current comp data from Levels.fyi, Glassdoor, and industry surveys through early 2026. It also covers which skills and credentials have a measurable impact on compensation.
Data Science Salary by Experience Level
The spread between entry-level and senior data science compensation is wider than almost any other tech role. Here's what the market looks like right now:
- Entry-level (0–2 years): $85,000–$105,000 base. Total comp including bonus typically lands around $95K–$115K outside of big tech.
- Mid-level (2–5 years): $115,000–$140,000 base. At this point, specialization matters — an ML engineer earns materially more than a general analyst.
- Senior (5–9 years): $140,000–$175,000 base. Stock compensation starts becoming a significant portion of total comp at this level.
- Staff / Principal (9+ years): $180,000–$250,000+ base at top-tier companies, with total comp frequently exceeding $350K at FAANG-tier firms.
These are base salaries. At public tech companies, RSUs can double or triple the effective annual compensation for senior and staff roles. A senior data scientist at Meta or Google with $160K base might be taking home $280K+ in total comp once stock vests.
Data Science Salary by City
Location still moves the number substantially, though remote work has compressed the gap compared to 2019–2021 levels.
- San Francisco / Bay Area: $130K–$165K base for mid-level. Highest absolute salaries, though cost of living eats most of the premium.
- New York City: $120K–$155K base. Finance-sector data science roles (quant roles at hedge funds) skew this market upward significantly.
- Seattle: $125K–$160K base. Amazon and Microsoft anchor the market; surrounding ecosystem follows their bands.
- Austin / Denver / Chicago: $100K–$130K base. Lower COL but comp has been rising sharply as remote candidates bid up local competition.
- Remote (US): Most companies have converged on "location-adjusted" pay. Google, Stripe, and Airbnb explicitly reduce base for non-hub locations by 10–25%.
If you're targeting maximum total comp and can stomach the cost of living, the Bay Area + Seattle corridor still wins on paper. If you're optimizing for purchasing power, Austin and Denver consistently show up well in surveys of working data scientists.
Data Science Salary by Specialization
The "data scientist" title covers a wide range of actual jobs. Specialization has a bigger impact on salary than most people realize:
- Machine Learning Engineer: $140K–$185K base nationally. The highest-paid adjacent role — combines software engineering rigor with model development. Demand is outpacing supply sharply post-2023.
- Data Scientist (core modeling): $115K–$155K base. Includes experimentation, feature engineering, production model deployment. A mature role now; competition is higher than 2018–2022.
- Data Analyst: $75K–$110K base. More SQL + BI tools, less Python + modeling. Often the entry point into the data science track.
- Business Intelligence Analyst: $80K–$115K base. Tableau/Power BI-heavy. Strong in finance and retail; undervalued in pure-tech environments.
- Data Engineer: $120K–$160K base. Pipelines, infrastructure, Spark, Kafka. Often earns more than data scientists at the same seniority level because the work is harder to offshore.
- AI/LLM Engineer: $150K–$200K+ base. The current hot end of the market. Demand is real but so is credential inflation — actual production experience commands a significant premium.
Which Industries Pay the Most
Your employer's industry is often more predictive of your salary ceiling than your individual skill level:
- Finance (hedge funds, prop trading, investment banks): Top of market. Senior quant/data science roles at Renaissance, Two Sigma, or Jane Street operate on a completely different pay scale — total comp in the $500K–$2M range is not unusual at the upper end.
- Big Tech (Meta, Google, Amazon, Apple, Microsoft): $250K–$400K+ total comp for senior roles. Base is competitive but stock is the story.
- Healthcare / Biotech: $110K–$150K base for experienced roles. Below pure tech but growing fast, particularly for ML applied to genomics and imaging.
- Retail / E-commerce: $100K–$135K base. Amazon is the outlier here. Brick-and-mortar retail lags significantly.
- Consulting / Advisory: $95K–$130K base at traditional firms (McKinsey, Deloitte), higher at boutique data-focused shops. Bonus structure varies heavily.
- Government / Nonprofit: $80K–$110K. Meaningful mission work; predictable comp growth; rarely competitive with private-sector ceiling.
Skills That Actually Move Your Data Science Salary
Not all skills are worth the same on the market. Based on current job posting analysis and compensation surveys, these have measurable salary impact:
- Python (core, not just notebooks): Table stakes. Expected at every level. Absence hurts you; presence alone doesn't differentiate.
- SQL at production level: Still underrated. Data scientists who can write clean, efficient SQL for analytical workloads are more productive than those who can't. Directly impacts your ability to work independently.
- Cloud platforms (AWS, GCP, Azure): $10K–$20K salary premium on average for roles that require production deployment. Certification helps at the entry level; portfolio projects matter more mid-career.
- MLOps / model deployment: Practitioners who can take a model from notebook to production API are worth more than those who can only develop. This is the most common skill gap on data science teams.
- Causal inference / experimentation design: Uncommon and valuable. A/B testing fluency is expected; causal inference beyond that (difference-in-differences, instrumental variables) commands a premium, especially at product-led tech companies.
- Domain depth: A data scientist who deeply understands the business domain (finance, healthcare, supply chain) is harder to replace than a generalist. Domain depth becomes more valuable than raw technical skill after year 5.
Top Courses to Build Skills That Pay
Credentials matter most at the entry level. After two years of work experience, your portfolio and demonstrated output matter far more than certificates. That said, structured learning accelerates the skill acquisition that leads to higher-bracket roles.
Introduction to Data Analytics (Coursera)
A rigorous foundation covering data collection, cleaning, analysis, and visualization — the core loop that data analysts and junior data scientists execute daily. Good starting point if you're transitioning from a non-technical background and want to validate your fundamentals before tackling Python or SQL at depth.
Tools for Data Science (Coursera)
Covers the actual working environment: Jupyter, RStudio, Git, Watson Studio. Not glamorous, but practitioners who can navigate the toolchain confidently get up to speed faster on real teams — which translates to earlier promotions and stronger performance reviews.
Python for Data Science, AI & Development by IBM (Coursera)
IBM's Python course is more production-oriented than most intro courses — it covers APIs, web scraping, and working with real datasets alongside the standard pandas/NumPy stack. If you want Python skills that transfer to actual job tasks rather than toy examples, this is the right starting point.
Analyze Data to Answer Questions (Coursera)
Part of the Google Data Analytics certificate track. Focuses on the full analysis workflow from hypothesis to insight communication — the skill set that directly determines whether a data analyst gets promoted to a data science track or stays in reporting roles.
Python Data Science (edX)
Covers statistical modeling, machine learning fundamentals, and data visualization using Python. Useful for analysts looking to move into data science roles where modeling experience is required for the salary jump to the $115K+ bracket.
Snowflake for Data Engineers (Udemy)
Snowflake is now the dominant cloud data warehouse in enterprise environments. Data engineers with Snowflake architecture experience are consistently among the highest-paid practitioners in the data space — this course covers the performance and design concepts that matter in production environments, not just the basics.
FAQ
What is the average data science salary in the US?
The BLS median is around $108K, but this figure includes a wide range of roles and geographies. Working data scientists in tech hubs typically earn $120K–$150K at the mid-level, with senior roles frequently exceeding $160K base. Total compensation (including stock and bonus) at large tech companies often runs 40–80% above base.
Is data science still worth it in 2026?
Yes, with caveats. The gold-rush era of 2018–2022 — where a bootcamp certificate could get you a $120K offer — is over. The market has matured. Entry-level is more competitive. But demand for experienced practitioners who can deploy models to production, work with LLMs, or apply causal inference remains strong, and compensation for those skills is still well above the median for knowledge workers.
How much do entry-level data scientists make?
Entry-level data scientists (0–2 years, typically with a relevant degree) earn $85K–$105K base outside of major tech hubs, and $100K–$125K in San Francisco, Seattle, or New York. Entry-level data analysts typically start $10K–$20K below those figures. The gap narrows after 2–3 years as skills and track records become the differentiator.
Do data science certifications increase salary?
At the entry level, yes — certifications signal baseline competency and get resumes past screeners. The Google Data Analytics certificate and IBM Data Science Professional Certificate on Coursera are the most recognized. Beyond the hiring threshold, certifications have minimal salary impact. Portfolio projects and work experience matter far more for the mid-to-senior level salary jumps.
What's the difference in salary between a data analyst and a data scientist?
Nationally, around $20K–$35K at comparable experience levels. Data analysts typically earn $75K–$110K; data scientists earn $95K–$140K at the same seniority. The gap reflects heavier modeling and coding demands for data science roles. The fastest path to data science compensation is building Python and ML skills while in an analyst role, then transitioning internally.
Which data science specialty pays the most?
Machine Learning Engineering and AI Engineering currently command the highest salaries in the data space — $140K–$200K+ base nationally, with strong upside at top-tier firms. Quantitative finance roles at hedge funds and prop trading firms pay even more, but the selection process is extremely competitive and requires deep mathematical background. For most practitioners, the ML Engineering track offers the best combination of high compensation and accessibility.
Bottom Line
Data science salary is highly variable — more so than most tech disciplines — because the title covers everything from Excel-heavy analyst roles to production ML infrastructure work. The single biggest lever on your comp is specialization: moving from "data analyst" to "data scientist" to "ML engineer" follows a clear upward trajectory, and each transition requires adding concrete technical skills, not just time on the job.
If you're entering the field, prioritize Python, SQL, and cloud fundamentals — these unlock the $95K–$110K entry bracket. If you're already working as an analyst and want to move into data science compensation territory, the gap is primarily about modeling experience and the ability to deploy code to production. The courses above address the specific skill gaps that show up most frequently in that transition.
The market is competitive but not saturated at the senior end. Practitioners who combine domain depth with solid engineering habits — clean code, version control, reproducible analysis — are still in strong demand and command compensation well above the BLS median.