From Data Analyst to Data Engineer: Consumer to Producer

Data analysts write queries against tables someone else built. Data engineers build the tables, the pipelines that feed them, and the infrastructure that keeps them reliable at scale. That distinction matters more now than it did two years ago. LLMs are getting remarkably good at the analytical side of data work: writing SQL, summarising datasets, generating dashboards. The engineering side, the pipelines that need to be observable, recoverable, and performant at 3am, remains squarely human.

Across the job postings we track, a third of the skills that data engineering roles ask for most frequently are already in data analyst job descriptions: SQL, Python, dbt, Snowflake, BigQuery. But the other two-thirds are almost entirely infrastructure. AWS shows up in 42% of data engineering roles but only 3% of analyst roles. Airflow appears in 39% versus 4%. Spark in 35% versus 5%. Kafka, the streaming layer, appears in 27% of engineering roles and essentially zero analyst postings. This is a different job, with a meaningful skill gap to bridge.

The transition from analyst to engineer is really a shift from consuming data infrastructure to building it. You go from writing queries against tables someone else set up, to being the person who designs how that data gets there, stays fresh, and scales when the business grows.

Where dbt sits in this transition

One skill is worth calling out specifically. dbt appears in 26% of data analyst roles and 36% of data engineering roles. It's the single tool that sits most naturally across both worlds: a bridge between the SQL-centric analyst workflow and the engineering-centric pipeline workflow. If you're an analyst who hasn't picked up dbt yet, it's the most efficient first step toward engineering because it extends what you already know (SQL and data modelling) into version-controlled, testable, production-grade transforms.

Analytics engineering: the halfway house

If the full leap to data engineering feels steep, analytics engineering is worth considering as an intermediate step, or as a destination in its own right. The role sits between analysis and engineering: you're still working primarily in SQL and dbt, but with more ownership of the data models, testing frameworks, and transformation pipelines that analysts downstream depend on.

The skill overlap between data analyst and analytics engineer is 8 out of 15 top skills, compared to 5 out of 15 for data engineering. SQL (96%), Python (65%), dbt (72%), Snowflake (49%), Looker (49%), Tableau (32%), BigQuery (23%), and Power BI (14%) all carry over directly. The bridging skills are narrower and less infrastructure-heavy: Airflow (35%), data modelling (19%), git (18%). You're learning version control and orchestration rather than cloud architecture and distributed computing.

There's a profile signal in the employer data too. 55% of analytics engineering roles come from scaleups, compared to 27% from enterprise companies. That's the inverse of almost every other data role we track. If you're drawn to smaller, faster-moving tech companies, analytics engineering is where those companies are hiring.

In the US, analytics engineering roles range from around $159k to $198k, sitting between data analyst ($129k to $165k) and data engineer ($166k to $230k) compensation. The volume is smaller (450 openings versus 1,570 for analysts and 2,000 for engineers), but the role is still growing.

The seniority picture

Data engineering roles skew senior. Across the openings we're tracking, 53% of data engineering roles are senior level and another 11% are staff or principal. Only 6% are junior. Compare that with data analyst roles where 15% are junior and 27% are mid-level.

This actually works in your favour if you're a mid-level analyst with a few years of experience. The market has very few entry-level data engineering openings, which means most employers expect to hire people who've already built context in a data role. Your analyst experience is an asset. The companies hiring data engineers know the role requires someone who understands the data before they start building pipelines around it.

What the salary data shows

In the US market, data engineering roles range from around $166k to $230k across the postings with published compensation. Data analyst roles sit lower, typically $129k to $165k. The exact numbers vary by city and company, but the direction is consistent across the openings we track.

Where the jobs are

Among the companies hiring data engineers in the openings we track, fintech is the largest single industry at 17% of roles, followed by professional services (17%) and consumer companies (10%). Data infrastructure companies, the ones building the tools data engineers use, account for 9% of openings. The mix is broader than for analyst roles, which concentrate more heavily in fintech and ecommerce.

There's also a volume signal. Globally, there are roughly 1.3 data engineering openings for every data analyst opening. In San Francisco that ratio stretches to 2:1, and for remote roles it's even higher at 2.2:1. If you're open to remote work, the supply and demand dynamics tilt further in favour of engineering roles.

Practical next steps

If you're making this move, here's how we'd sequence it based on what the data shows.

Start with what extends your existing skills. dbt is the obvious first step if you haven't used it already, because it takes SQL you already write and puts it in a production context with version control, testing, and documentation. Git and basic CI/CD practices come naturally alongside this. If analytics engineering appeals as an intermediate step, you could stop here and already be competitive for those roles.

For cloud infrastructure, the reality is that most analysts will already be working within whatever cloud provider their employer uses. If that's AWS (which appears in 42% of data engineering roles), go deeper on S3, Glue, Redshift, and IAM. If it's GCP (20%), focus on BigQuery, Dataflow, and Cloud Storage. The concepts transfer between platforms, so depth on one matters more than surface familiarity with both. If you're between jobs or building skills independently, AWS has the broadest market share.

Learn an orchestration tool. Airflow is still the market leader at 39% of postings, but Dagster and Prefect are growing. The important thing is understanding DAGs, scheduling, monitoring, and failure handling. The specific tool is secondary to the mental model.

Finally, build something. The gap between knowing what Kafka does and having used it to move data through a pipeline is where interviews separate analysts from engineers. A side project that ingests, transforms, and serves data through a real pipeline will demonstrate engineering thinking more effectively than any certification.

You already have the data literacy, the SQL fluency, and probably some Python. The engineering skills are learnable, the market is hiring for them at a higher rate and a higher price point, and the resulting role is better insulated against the AI-driven automation that's reshaping the analytical side of data work. Whether you go directly to data engineering or take the analytics engineering stepping stone, the direction is clear.

---

Methodology: This analysis draws on approximately 5,400 job postings collected between November 2025 and February 2026 from company career pages. Postings span London, New York, San Francisco, Denver, and Singapore. Skills were extracted from full job descriptions using an LLM classifier. Salary data is limited to US markets where disclosure is more common. Recruitment agency listings and out-of-scope roles were excluded. For skills analysis, the data analyst sample is 214 postings and the data engineer sample is 453 postings. Analytics engineer analysis is based on 170 postings. Salary sample sizes: data engineer n=107, analytics engineer n=50, data analyst n=41.