Today marks a massive milestone for our company as we launch Extract, our brand-new high-performance data movement platform.
We designed Extract to be the world’s most efficient data pipeline. Powered by Rust, Extract offers lightning-fast, flexible data movement with an intuitive UI, powerful APIs, and incredible transparency—all at ridiculously affordable rates.
Extract builds on our decade of experience from Singular, the G2-rated world’s best mobile attribution platform, working with global clients like Microsoft, Uber, Nike, DoorDash, Instacart, and Shopify. At Singular, we handle data in every form—real-time, batch, incoming, and outgoing, and manage thousands of integrations with third-party systems in an insanely fast-moving ecosystem.
With AI dominating most companies’ tech stack, the amount of data transfer and processing has grown tenfold in the last 24 months. This requires a radical rethinking of the tech stack towards efficiency that can sustain that.
To date, companies choose between two options:
- Complex: Building their own pipelines or deploying some open-source connector, spending enormous time maintaining it, and mostly reinventing the wheel.
- Expensive: Cloud-based ELT platform that is easy to start with but quickly becomes prohibitively expensive.
We set out to give you a third option: easy & affordable.
Let me tell you a little bit about what you can expect.
Radically efficient
Almost all existing ELT tools are based on Python or Java as the language of choice. Those languages made a lot of sense a decade ago, and when a company is hard-pressed to release a product to market—it’s the fastest path (to be fair, that’s what Singular was built on, too).
With Extract, we took our time to build the most optimized foundation we could. That’s why we chose Rust, which has zero runtime, no garbage collection, and C/C++-like performance without memory bugs. We then built on that with an obsessive focus on lowering resource consumption and increasing our compute and memory utilization.
Data moving products require hundreds (if not thousands) of integrations to be useful. We decided to build and maintain all our connectors in-house. While we love open-source, there isn’t a reliable community of open-source connectors written in Rust, certainly not ones with our strict resource requirements and logging standards.
The result? A super efficient platform that outperforms every other product in the category. And those cost savings are directly passed on to you, the customer.
Transparent & customizable
Cloud ELTs and data movement platforms have been slow to innovate. Some have completely frozen in time (mostly after acquisitions), while others have been iterating at a snail’s pace.
We designed the system to be something we, as data engineers, would love to use. That meant that apart from being affordable and fast, it had to be fun. We put a huge emphasis on transparency because, as an engineer, seeing a loading spinner without knowing what’s going on is quite offensive.
The system also offers a lot of observability. You will be able to see logs on almost everything—from when you simply “log in” to a data connector, to the full-blown execution log of your pipelines. All logs update in real time, show up in the product, and have unlimited retention. So, if you ever want to investigate something, you have a great interface with advanced filters to do that.
In addition to the extreme transparency, we made sure the product was highly customizable. You can control:
- When we extract data: including presets, specific times and crontab expressions;
- How we extract data: Full refresh, Incremental Syncs, Partition, CDC;
- How we load data: Replace, Append, Upsert, Soft-Delete, History Mode, Data Contracts; and
- How we manage schema evolution
You can even make multiple configuration sets for the same Source+Destination pair via our Connection objects.
Stress-tested & trusted
While this is our first public announcement, we’ve onboarded over 50 customers to the platform in the past year, many of which moved from other ELT providers, and the feedback has been incredible.
We are also in the process of migrating our ETL capabilities from Singular to the Extract experience, which will boost Extract to 1000+ customers this year.
We also evaluated Salesforce’s Datorama, but Extract kept it much more simple, more easy. We just need the data to be pulled, customizable, and I can do that without the need of a Data Engineer”
Gal Karniel
Product Director @ SciPlay
It’s working very well for us because of the flexibility and the scalability. We can easily toggle for what data we want and shoot it to our warehouse, which powers 10 more of our most important internal reports”
Liri Rozenthal
BI Developer @ Miniclip
Get started today
We think you’ll love the Extract experience. It’s easy to start and completely free.
As a matter of fact, our free tier offers 1 million monthly rows, unlimited users, unlimited sources, and destinations!
Our first paid tier starts at $15/month and gives you complete flexibility if you have over a million monthly rows. From there, you get cost efficiencies based on volume and some additional capabilities (more frequent syncs, SSO, etc).
Check out all our pricing here.
I’d really like you to experience the future of data pipelines. Extract offers a seamless, efficient, and cost-effective solution designed to scale your business, and I think we’re making it super-easy to just try it out. Whether you’re a data engineer, analyst, or part of a larger BI team, our platform is tailored to fit your needs.
We’re excited for you to experience Extract firsthand and see why it’s set to become the preferred choice for modern data teams. Stay tuned for more updates as we continue to expand our capabilities and redefine what’s possible in data integration.