Reverse ETL vendor Census raises $60M

Data startup Census raised $60 million in a Series B round of funding, bringing total funding to date to $80.3 million.

Census, based  in San Francisco, was founded in 2018 and develops what is known as a reverse ETL platform.

With regular ETL (extract, transform and load), data is loaded into a data warehouse, where it can be used for data analytics. With reverse ETL, the basic idea is to enable data warehouse information to be used in other systems, including customer data platforms, operations and business intelligence.

Census competes in the reverse ETL market with Hightouch, Omnata and others.

In this Q&A, Boris Jabes, CEO and co-founder of Census, outlines the opportunities for reverse ETL and what the future holds for the vendor.

Why did you start Census as a reverse ETL vendor and why are you now raising money?

Boris Jabes

Boris Jabes: To me, it seemed obvious three and a half years ago when we started Census that customer data, and any important company data that you have, shouldn't be piecemeal in different places. Rather, it should be centralized somewhere and then federated to every application whenever you want, and the data should be all the same. You can only achieve that if you have kind of a hub-and-spoke data model.

It became obvious to us that the correct central layer for the data was the emerging cloud data warehouse, which is a concept that is still on the rise. We started with one connector from Amazon Redshift to Salesforce back in 2018 and we've expanded in the years since then to support more warehouses and more applications.

We're raising money now because we were just crumbling under the growth this past year. We were just struggling to keep up with the demand. So, that's the first and main reason why we're raising money; we just need to be able to grow our team to keep up with the demand.

Reverse ETL has become the generic word for what our tools do. I would say that technically it's a misnomer because ETL has no direction.

How do you define reverse ETL at Census?

Jabes: Reverse ETL has become the generic word for what our tools do. I would say that technically it's a misnomer because ETL has no direction.

The reason people gravitated to the term reverse ETL is we now have a modern kind of data stack and the stack starts with ingestion, often into a data warehouse. What we built back in 2018 was a way to publish the data back out to applications tools.  The market just kind of started calling this reverse ETL. So that's the term, but I would say on its own, it doesn't describe really anything.

I think the best way to think about it is that reverse ETL describes how the data moves; it moves from the data warehouse into applications.

So I think that is how our product does what it does. You don't have to hire an engineer to figure out how to convert data types from your data warehouse into Marketo or Salesforce or any other business application. That's what our product does.

That still really doesn't describe what the point of all this is for us. The goal from the beginning was the realization that really amazing data is in every company and it  should be spread across every corner of the company to get maximum impact from it.

Our view is that a data warehouse dashboard is not enough. You should put it into the tools where people do their work.

What's next for Census and reverse ETL?

Jabes: In terms of what what's going to come next, there's actually a lot of things.

The way I think about it is if you're going to operationalize your data, you need a really good membrane between the data team and the rest of the business team, which is what Census is. So how do we increase trust in your data? How do we increase reliability? How do we increase the scale and how do we bridge the gap between data and ops teams? So those are a lot of the things that we're going to work on in the coming year or two years.