Data scientists in the United States earn a median base salary of approximately $108,000 according to the Bureau of Labor Statistics 2023 Occupational Employment and Wage Statistics, but that single number obscures an enormous range. A junior data analyst at a regional insurance company and a senior ML researcher at Meta both carry the title "data scientist," yet their total compensation can differ by a factor of five or more. The key variables that drive this spread -- seniority, geography, industry, company tier, and the structure of total compensation -- are what you actually need to understand before negotiating your next offer or planning a career transition.
The data science salary landscape is also unusually opaque compared to fields like law or medicine, where compensation follows predictable guild structures. Data science sits at the intersection of academia, engineering, and business analysis, pulling pay norms from all three worlds. That creates enormous variance, but it also creates opportunity: candidates who understand how the market is structured consistently outperform those who anchor on misleading averages from generic salary sites.
This article draws on Bureau of Labor Statistics (BLS) data, Glassdoor's 2024 salary reports, Levels.fyi crowd-sourced compensation data, and the 2024 Stack Overflow Developer Survey to give you a grounded, level-by-level picture of what data scientists actually earn across major markets -- plus a breakdown of how total compensation packages work at different company tiers.
"Salary data for data scientists is almost always an underestimate of what top performers take home, because the crowd-sourced data skews toward people who are dissatisfied or recently hired. The ceiling is considerably higher than the median." -- Eugene Yan, applied scientist and author of the ApplyingML newsletter, 2023
Key Definitions
Base salary: The fixed annual cash payment before bonuses or equity. This is what most public salary surveys report and is almost always the floor of total compensation at technology companies.
Total compensation (TC): The sum of base salary, annual performance bonus, and annualized equity (RSUs or options). At large tech companies, TC can be 1.5x to 3x base salary -- making base salary a misleading comparison metric when evaluating offers across company types.
RSU (Restricted Stock Unit): A grant of company shares that vest over time, typically on a four-year schedule with a one-year cliff. The value fluctuates with stock price, which creates both upside (if the stock appreciates) and risk (if it declines).
Levels.fyi: A crowd-sourced compensation database primarily covering US tech companies. Submissions are verified by offer letters, making it more reliable than anonymous surveys for understanding senior-level compensation structures.
BLS Occupational Employment and Wage Statistics (OEWS): The US government's official wage survey, published annually. It captures broad occupational categories and is the most methodologically rigorous source, though it lags real-time market conditions by 12-18 months and does not capture equity compensation.
US Data Scientist Salaries by Level
| Level | Experience | Base Salary (US) | Total Comp (Big Tech) | Total Comp (Other Industries) |
|---|---|---|---|---|
| Entry-level | 0-2 years | $75,000 - $110,000 | $140,000 - $200,000 | $90,000 - $130,000 |
| Mid-level | 3-6 years | $120,000 - $160,000 | $200,000 - $280,000 | $130,000 - $170,000 |
| Senior | 7+ years | $160,000 - $220,000 | $300,000 - $500,000 | $170,000 - $220,000 |
| Staff / Principal | 10+ years | $200,000 - $280,000 | $400,000 - $700,000+ | $200,000 - $280,000 |
The BLS reported a median annual wage of $108,020 for "data scientists" in its 2023 OEWS release, placing the occupation in the top 10% of all US jobs by pay. However, this figure captures an enormous range across levels, industries, and company types. The 10th percentile earned $59,500 and the 90th percentile earned $184,600 -- a spread of over 3x that reflects the diversity of roles grouped under the "data scientist" umbrella.
Entry-Level (0-2 Years Experience)
Entry-level data scientists -- typically those finishing a graduate degree or a structured bootcamp with one meaningful portfolio project -- earn between $75,000 and $110,000 in base salary depending on location and industry. According to Glassdoor's 2024 US data, the median entry-level data scientist base is $95,000, with total compensation ranging from $90,000 to $130,000 when bonuses are included.
At large tech companies (the FAANG tier), the picture changes substantially. Entry-level L4/E4-equivalent data scientists start at $140,000 to $180,000 in base, with total compensation often reaching $200,000 or more when RSU grants are factored in over the first year's vesting period. A concrete example: a 2024 Levels.fyi entry for a new-grad data scientist at Google shows a base of $150,000, a signing bonus of $30,000, and an annual RSU vesting of approximately $50,000 -- yielding first-year total compensation of roughly $230,000.
Entry-level candidates for FAANG roles are expected to pass rigorous technical screens involving SQL, probability, machine learning theory, and system design -- a significantly higher bar than most other industry roles. The compensation premium reflects both the difficulty of the interview process and the revenue these companies generate per employee.
Mid-Level (3-6 Years Experience)
Mid-level data scientists commanding independent project ownership and some cross-functional influence typically earn $120,000 to $160,000 in base salary in the US. Total compensation at this level at top-tier tech companies commonly lands between $200,000 and $280,000 according to Levels.fyi 2024 data for L5/E5 equivalents.
Outside of big tech, mid-level salaries are meaningfully lower. Finance, consulting, and healthcare companies pay $130,000 to $170,000 in total compensation at this level, with less equity but often more stability and clearer domain expertise paths.
The compensation gap between big tech and other industries is most pronounced at mid-level -- the point where equity grants begin to diverge significantly. A mid-level data scientist at a Series B startup might receive meaningful equity in terms of percentage ownership, but the expected value of that equity is highly uncertain compared to publicly traded RSUs at an established tech company.
This is also the level where specialization begins to matter for compensation. Data scientists who develop deep expertise in high-demand areas -- natural language processing, recommendation systems, causal inference, or ML infrastructure -- command premiums of 10-20% over generalists at the same level, according to analysis of Hired's 2024 State of Tech Salaries report.
Senior Level (7+ Years Experience)
Senior data scientists -- those leading project strategy, mentoring junior staff, and owning technical decisions across multiple initiatives -- earn $160,000 to $220,000 in base salary across the US market. Levels.fyi data for senior data scientists at Google, Meta, and Amazon shows total compensation packages of $300,000 to $500,000 at L6/E6 level, driven heavily by stock grants that can represent 50-60% of total compensation.
At this level, the distinction between data scientist and applied scientist becomes important. Companies like Amazon and Meta use "applied scientist" titles for roles that are more research-oriented, and these roles often command higher compensation than equivalently leveled "data scientist" positions that are more analytics-focused.
Staff and Principal Level (10+ Years)
At director, principal, or staff scientist levels, total compensation at elite tech firms routinely exceeds $500,000 annually and can reach $700,000 or more. These roles are rare and typically require both deep technical expertise and demonstrated organizational impact -- the ability to influence research direction, shape technical strategy, and mentor multiple teams simultaneously.
A 2024 Levels.fyi entry for a staff-level data scientist at Meta shows base salary of $250,000, bonus of $75,000, and annual RSU vesting of approximately $375,000 -- total compensation of $700,000. These numbers are real but represent the top of the market; the median staff-level data scientist at a non-FAANG company earns closer to $250,000-$350,000 in total compensation.
Regional Salary Comparison
| Region | Entry-Level (TC) | Mid-Level (TC) | Senior Level (TC) |
|---|---|---|---|
| US (Big Tech, Bay Area) | $140,000 - $200,000 | $200,000 - $280,000 | $300,000 - $500,000 |
| US (General Market) | $90,000 - $130,000 | $130,000 - $170,000 | $170,000 - $220,000 |
| UK (London) | GBP 35,000 - 50,000 | GBP 55,000 - 75,000 | GBP 80,000 - 110,000 |
| Germany (Berlin/Munich) | EUR 45,000 - 60,000 | EUR 60,000 - 90,000 | EUR 85,000 - 130,000 |
| Netherlands | EUR 40,000 - 55,000 | EUR 55,000 - 85,000 | EUR 80,000 - 120,000 |
| India (Major Cities) | INR 600,000 - 1,200,000 | INR 1,500,000 - 3,000,000 | INR 3,500,000 - 8,000,000 |
| Canada (Toronto/Vancouver) | CAD 65,000 - 90,000 | CAD 90,000 - 130,000 | CAD 130,000 - 180,000 |
United Kingdom
UK data scientist salaries lag US figures significantly in absolute terms, though purchasing power parity closes the gap somewhat. Glassdoor UK data from 2024 shows entry-level GBP 35,000-50,000 (approximately $44,000-$63,000), mid-level GBP 55,000-75,000 (approximately $69,000-$95,000), and senior GBP 80,000-110,000 (approximately $101,000-$139,000).
London adds a roughly 10-15% premium over the national median. The largest tech employers in the UK (Google London, Amazon UK, DeepMind) pay closer to European band structures than US structures, though some US multinationals apply a "UK band" that narrows the gap for senior hires. DeepMind, in particular, is an outlier -- senior research scientists there earn packages competitive with Bay Area rates, reflecting the scarcity of top AI research talent globally.
The UK market is also notably smaller in volume. The number of active data scientist job postings in the UK is roughly one-fifth of the US market, which creates both more competition at entry level and stronger negotiating power for senior specialists who can demonstrate rare expertise.
European Union
EU salaries vary significantly by country. Germany is the strongest EU market for data scientists by volume and compensation. Berlin has a growing ML startup ecosystem (with companies like DeliveryHero, Zalando, and numerous AI startups), while Munich's engineering-heavy corporate culture (BMW, Siemens, Allianz) provides stable mid-range employment.
Equity compensation is far less common in EU roles than in US tech companies, making total compensation comparisons less favorable for EU-based positions than base salary comparisons suggest. A German data scientist earning EUR 90,000 in base with no equity is earning meaningfully less in total compensation than a US counterpart earning $130,000 with $40,000 in annual RSU vesting. This structural difference is the primary reason the US-EU compensation gap is larger than raw base salary numbers indicate.
The Netherlands has emerged as a secondary hub, particularly Amsterdam, driven by companies like Booking.com, Adyen, and the European offices of US tech firms. Dutch salaries are comparable to German ones, with slightly better tax treatment through the 30% ruling for international hires.
India
India's data science job market has grown rapidly but operates on a fundamentally different compensation scale. Per Glassdoor India and AmbitionBox 2024 data, entry-level earns INR 600,000-1,200,000 (approximately $7,200-$14,500), mid-level earns INR 1,500,000-3,000,000 (approximately $18,000-$36,000), and senior/lead earns INR 3,500,000-8,000,000 (approximately $42,000-$96,000).
Indian operations of US multinationals (Google India, Microsoft India, Flipkart) pay at the top of these ranges and sometimes extend equity tied to US parent company stock. A senior data scientist at Google India might earn INR 5,000,000-8,000,000 in base plus significant RSUs, making total compensation highly competitive by Indian standards. The cost of living differential is significant -- purchasing power in major Indian cities makes these figures more competitive in context than the dollar conversions suggest.
Data Scientist vs. ML Engineer vs. Data Analyst: Pay Differences
The three most commonly confused roles in the data ecosystem have meaningfully different compensation profiles:
Data analyst roles primarily use SQL, Excel, and BI tools to answer defined business questions. US median base salary is approximately $75,000-$85,000 according to BLS 2023 data, with senior analysts reaching $110,000-$130,000 at tech companies. The skill bar is lower, and the ceiling is correspondingly lower -- but the path to entry is also more accessible, and demand is strong.
Data scientists occupy the middle ground, commanding a premium over analysts for their statistical modeling, machine learning, and Python/R proficiency. US median is around $108,000 base, with wide variance by level and company.
ML engineers -- who productionize models, build pipelines, and work closely with software infrastructure -- earn at parity with or slightly above data scientists at mid and senior levels. Levels.fyi 2024 data shows mid-level ML engineers at major tech companies earning $220,000-$320,000 in total compensation. The software engineering component of the ML engineer role means they compete for the same compensation bands as senior software engineers at many companies.
| Role | US Median Base | Big Tech TC (Mid-Level) | Key Skills |
|---|---|---|---|
| Data Analyst | $75,000 - $85,000 | $120,000 - $160,000 | SQL, Excel, BI tools, business acumen |
| Data Scientist | $108,000 | $200,000 - $280,000 | Python/R, statistics, ML, experimentation |
| ML Engineer | $115,000 - $130,000 | $220,000 - $320,000 | Python, MLOps, distributed systems, production ML |
| Data Engineer | $110,000 - $125,000 | $200,000 - $280,000 | SQL, Spark, Airflow, cloud infrastructure |
The career trajectory increasingly favors specialization. Generalist "full-stack data scientist" roles are declining in number at large companies, replaced by more specialized positions: experimentation/causal inference, ML engineering, data engineering, and analytics engineering. Each specialization has its own compensation trajectory, and the highest-paying paths tend to be those closest to either core ML research or production engineering.
How Total Compensation Packages Are Structured
Understanding your total compensation requires looking beyond the headline number to each component's characteristics:
Base salary is the predictable cash floor. It determines your tax bracket, your mortgage qualification, and your baseline financial stability. It is also the most portable number -- you can negotiate based on it at any company. Most companies cap base salary at a certain level (Google, for instance, rarely exceeds $300,000 in base for any individual contributor), with additional compensation delivered through equity.
Annual bonus typically runs 10-15% of base at mid-size companies and 15-25% at financial services firms. At tech companies, bonuses are often smaller in percentage terms (10-20%) but are supplemented by equity. Bonuses are usually tied to company performance, team performance, and individual ratings -- meaning they are not guaranteed. In a bad year, your bonus might be 50-70% of target; in a strong year, it might be 120-150%.
Equity (RSUs at public companies) vests over four years typically with a one-year cliff. A $200,000 RSU grant that vests over four years adds $50,000 per year in pre-tax compensation assuming the stock price holds. At fast-growing companies, RSU grants can appreciate significantly -- a data scientist who joined Meta in early 2023 with a standard RSU package saw their equity value increase by over 150% through early 2025 as the stock price rose from approximately $130 to over $600. Conversely, those who joined in 2021 at peak stock prices experienced the opposite.
Sign-on bonuses are used to bridge equity vesting gaps when switching employers. A company trying to recruit someone away from unvested RSUs may offer a $50,000-$100,000 sign-on to compensate for what is left on the table. These are non-recurring and should not be counted in your running annual compensation figure, but they are a legitimate and negotiable part of the offer.
Benefits -- health insurance, 401(k) matching, employer HSA contributions, and paid parental leave -- have real monetary value that varies dramatically between employers. A tech company with fully covered family health insurance (worth roughly $25,000-$30,000 annually) and a 4% 401(k) match is paying several thousand dollars more in effective compensation than a company where you cover your own premiums. The 2024 Glassdoor survey found that 80% of employees would prefer additional benefits over a pay raise, though this preference varies by career stage and financial situation.
Industry Pay Differences
Not all data science jobs pay equally, even at the same level and experience:
Highest-paying industries: Technology (especially FAANG and AI-focused companies), quantitative finance (hedge funds, prop trading firms, investment banks), and defense/intelligence contracting. Compensation at top-tier hedge funds like Two Sigma, DE Shaw, Citadel, and Jane Street for quant data scientists can exceed $500,000-$1,000,000 in total compensation at senior levels, exceeding even FAANG rates. These firms compete directly with big tech for the same talent pool.
Mid-range industries: Consulting firms (McKinsey QuantumBlack, BCG Gamma), SaaS companies outside the top tier, e-commerce, digital media, and pharmaceutical companies. These industries offer competitive compensation with often better work-life balance and clearer paths to domain expertise.
Lower-paying industries: Government agencies, nonprofits, traditional retail, academic institutions, and regional healthcare systems. The spread between top and bottom industry is roughly 2x to 3x at the senior level -- a senior data scientist at a hedge fund may take home $400,000 total compensation while a counterpart at a state government agency earns $130,000.
The industry pay gap is growing, not shrinking. The McKinsey Global Institute's 2023 report on the analytics workforce found that AI-specialized roles command premiums of 25-40% over equivalent non-AI data science positions, and this premium has increased every year since 2020 as demand for generative AI expertise has surged.
The Impact of AI on Data Science Compensation
The rise of generative AI and large language models since 2022 has created a significant shift in the data science job market. Roles focused on LLM fine-tuning, retrieval-augmented generation (RAG), and AI safety have seen compensation increases of 20-35% compared to traditional data science roles at the same level, according to Hired's 2024 salary data.
At the same time, some traditional data science tasks -- exploratory data analysis, basic modeling, and report generation -- are increasingly automated by AI tools. This has compressed compensation for purely analytical roles while inflating it for roles that require building, deploying, and evaluating AI systems. The data scientists best positioned for continued compensation growth are those who combine deep ML expertise with software engineering skills -- the ability to build production systems, not just notebooks.
Practical Salary Negotiation Strategies
Before your next negotiation, know which data points are most relevant to your situation:
Use BLS data to establish a credible floor. The OEWS median is conservative but defensible as an anchor in salary discussions with companies that may not be familiar with tech-specific compensation.
Use Glassdoor for industry and city-specific benchmarks. Filter by company size and read the methodology notes to understand what the figures include.
Use Levels.fyi for tech company negotiations specifically. The data is the most reliable for understanding total compensation structures at public tech companies and is far more granular than general salary surveys.
Always negotiate total compensation, not just base. At a company that offers equity, asking for more RSUs rather than higher base can be more tax-efficient and more palatable to hiring managers who have constrained base salary budgets. A $10,000 increase in annual RSU value may be easier to secure than a $10,000 base salary increase.
Understand vesting schedules before accepting. A large RSU grant with a one-year cliff means you cannot leave without forfeiting unvested equity for at least twelve months -- price that constraint into your decision. Some companies (notably Amazon) backload their vesting schedules: only 5% vests in year one and 15% in year two, with 40% in each of years three and four. This dramatically affects your effective compensation in the first two years.
Time your job search strategically. Compensation offers are strongest when companies are actively competing for talent. The strongest negotiating positions come from having multiple offers simultaneously -- a practice that requires disciplined job search timing but can increase total compensation by 15-25% compared to negotiating a single offer in isolation.
The geography premium for US tech work is real and large. If remote work is available, taking a US-domiciled role from another country may be the highest-leverage compensation move available to international candidates. However, be aware that many US companies are moving toward location-based pay bands that adjust compensation downward for employees outside of high-cost metropolitan areas.
References and Further Reading
- Bureau of Labor Statistics. (2024). Occupational Employment and Wage Statistics: Data Scientists. US Department of Labor. bls.gov/oes
- Glassdoor. (2024). Data Scientist Salaries. Glassdoor Economic Research.
- Levels.fyi. (2024). Data Scientist Compensation Data. levels.fyi/t/data-scientist
- LinkedIn Salary Insights. (2024). Data Scientist Median Salary by Location. linkedin.com
- Glassdoor UK. (2024). Data Scientist Salary in United Kingdom. glassdoor.co.uk
- AmbitionBox. (2024). Data Scientist Salaries in India. ambitionbox.com
- Yan, E. (2023). ApplyingML Newsletter: Career and Compensation Edition. eugeneyan.com
- ONET Online. (2024). Data Scientists: Summary Report. National Center for ONET Development.
- Hired. (2024). State of Tech Salaries Report. hired.com
- McKinsey Global Institute. (2023). The Age of Analytics: Competing in a Data-Driven World. mckinsey.com
- Stack Overflow. (2024). Developer Survey: Salary and Compensation Section. survey.stackoverflow.co/2024
- Radford (Aon). (2024). Technology Industry Compensation Survey. aon.com/radford
- Burtch Works. (2024). Data Science and Analytics Salary Survey. burtchworks.com
- Federal Reserve Bank of New York. (2024). Labor Market Conditions for STEM Workers. newyorkfed.org
Frequently Asked Questions
What is the average data scientist salary in the US?
The BLS median annual wage is around \(108,000, but total compensation at major tech firms often exceeds \)180,000-$280,000 when stock and bonuses are included. The figure varies enormously by level and company type.
Do data scientists earn more than software engineers?
At mid-level, software engineers typically earn slightly more. At senior and staff levels, data scientists specialising in ML research can match or exceed SWE compensation, particularly at AI-focused companies.
How does data scientist pay differ between the US and UK?
US data scientists earn roughly 40-60% more in base salary than UK counterparts, primarily because US tech salaries are globally exceptional rather than UK salaries being low. The gap is largest at senior levels with equity.
What is total compensation vs base salary for a data scientist?
Total compensation includes base salary, annual bonus (typically 10-20% of base), and equity (RSUs). At large tech companies, equity can double or triple base salary over a four-year vest, making it the largest compensation component.
Which companies pay data scientists the most?
Meta, Google, Apple, Microsoft, and Amazon consistently offer the highest total compensation at \(200,000-\)500,000+ at senior levels per Levels.fyi data. Top quantitative hedge funds can exceed even FAANG rates for specialised roles.
