Matillion is the cloud-native middle ground in the ETL market. Founded in 2011 in Manchester, UK by Matthew Scullion and Ed Thompson, Matillion built a visual, GUI-driven ETL platform specifically for cloud data warehouses — Redshift first, then Snowflake, BigQuery, Databricks, and Synapse. Where Informatica is the heavyweight legacy and dbt is the code-first modern, Matillion is the drag-and-drop product for teams that want a visual interface but also want to live inside a cloud warehouse.

Plain English: Matillion is what you'd get if you took Informatica PowerCenter, threw away the on-prem server, and rewrote it to push every operation down into Snowflake. You build pipelines by dragging components onto a canvas. Behind the scenes, Matillion generates SQL and runs it against your warehouse, so the heavy compute happens where it should. It's the visual ETL experience without the on-premise overhead.

Origin Story: A Manchester Bet on Redshift

Matillion started as a UK consultancy doing business intelligence work in the late 2000s. Matthew Scullion and Ed Thompson saw two things happening simultaneously: enterprise customers were tired of the cost and complexity of on-prem ETL tools, and AWS Redshift had just launched (2012) as the first credible cloud data warehouse. They made a contrarian product bet for a UK company: build a GUI-based ETL tool exclusively for Redshift, sell it through the AWS Marketplace, and deploy it as an EC2 AMI.

This was a niche bet in 2014. Redshift was not yet dominant. Snowflake had not yet launched publicly. Most enterprise data integration spend was still flowing to Informatica and IBM. But it was the right niche at the right time. As Redshift grew, Matillion grew with it — specifically targeting the customer who wanted a Redshift-native tool, didn't want to write code, and didn't want to install Informatica.

By 2017, Matillion had ported the product to Snowflake (which became their largest segment), then BigQuery, then Synapse, then Databricks. They raised a $100M Series D in 2021 led by General Atlantic, and a $150M Series E later that same year valuing the company at $1.5B. In 2023 they rebranded the umbrella product as the Matillion Data Productivity Cloud — a unified SaaS platform combining ingestion, transformation, orchestration, and (more recently) AI features.

What's Actually in the Box

A Matillion job is a visual canvas of components. Components fall into two buckets: Orchestration (control flow — run jobs, run scripts, conditionally branch) and Transformation (the actual SQL operations — joins, filters, aggregates, calculator columns, lookups). When you run a transformation job, Matillion compiles the canvas into SQL and submits it to your warehouse. Your warehouse does the work.

This is the architecturally clever part: Matillion is mostly a code generator. The Matillion server (or SaaS instance) is small and stateless. It exists to design pipelines, schedule them, monitor them, and emit SQL. The actual compute happens on Snowflake/BigQuery/Databricks. You pay your warehouse for compute regardless, so Matillion's overhead is the design surface, not the execution layer.

The product also includes:

• Pre-built connectors for ~150 SaaS sources (Salesforce, HubSpot, NetSuite, Marketo, Google Analytics, etc.) similar to but smaller than Fivetran's catalog.
• CDC connectors for relational databases.
• Git integration for version-controlling jobs (relatively new and still maturing compared to dbt's Git-native model).
• Matillion AI / Maia — a natural-language assistant for generating jobs and components, launched as part of the broader AI push in 2024.
• dbt integration — Matillion can run dbt projects as part of its orchestration, acknowledging that some teams want both.
• Multiple deployment models — the original AMI/VM deploy is still available; the SaaS Data Productivity Cloud is the recommended path.

Where Matillion Fits in the Market

Matillion's competitive position is the most interesting — and the most precarious — in the modern data integration market. The company sits between three forces, each pulling in a different direction:

The Fivetran + dbt camp. Most modern data teams don't want a visual ETL tool at all. They want managed connectors (Fivetran) and SQL-based transformations in Git (dbt). Matillion's visual canvas is, to these teams, an anti-feature — a fragile abstraction over SQL that makes code review harder, version control messier, and AI-assisted development more awkward. The "code-first" zeitgeist is the strongest secular trend in the modern data stack, and it works against Matillion.

The Informatica camp. Large enterprise IT organizations love visual ETL tools, because they let non-coding business analysts build pipelines, they integrate with existing GUI-driven workflows, and they look great in procurement demos. Matillion can win these deals against Informatica by being cloud-native, simpler, and substantially cheaper. But these buyers are slow to switch, and Informatica has a 30-year head start on the relationships.

The warehouse-native camp. Snowflake's Snowpark, Snowflake Dynamic Tables, Databricks Lakeflow, and BigQuery Dataform are all incursions into the territory Matillion serves. If your transformation tool ships natively with your warehouse, why pay a third party to wrap it?

Matillion's response to all of this has been to lean into the cloud-native visual experience and invest heavily in AI-assisted development. The bet is that the next wave of data team hiring will include analysts who don't write SQL fluently but can describe what they want to an AI, and that a visual canvas is a better human interface for reviewing AI-generated transformations than a wall of code. It's a plausible bet, but the jury is out.

The Honest Take

Matillion is a genuinely well-engineered product, and for the right buyer it is an excellent choice. The right buyer, in our view, looks like this:

• Mid-market or enterprise with a small data team
• Snowflake or BigQuery already in production
• A mix of SQL-fluent and non-SQL-fluent contributors who need a shared workspace
• Allergy to managing dbt + Airflow + Fivetran + monitoring tools as separate products
• Comfort with vendor lock-in in exchange for a unified UX

For this buyer, Matillion is more pleasant than assembling the modern data stack from parts. For everyone else — particularly engineering-led teams at modern companies — Matillion is a category mismatch. The market has decisively moved toward unbundled, code-first, Git-native tooling, and Matillion is fighting that current.

The encouraging thing for Matillion is that the cloud data warehouse market is still growing fast, the long tail of mid-market customers is enormous, and a lot of those customers will never adopt dbt for cultural reasons. The discouraging thing is that "visual ETL for the cloud warehouse" is a shrinking premium niche, not an expanding one.