Ashby Selects Polytomic for High-Scale PostgreSQL CDC

About Pathpoint

Ashby is a modern all-in-one recruiting platform encompassing applicant tracking, interview scheduling, feedback management, candidate relationship management, and recruiting analytics.

Some of the world's fastest-growing companies run on Ashby: from public companies like Snowflake and Shopify to hypergrowth startups like Notion and Ramp.

We spoke to Kevin Connolly, Head of Data and Analytics at Ashby, about their use of Polytomic to run high-scale PostgreSQL CDC (Change Data Capture) into Snowflake and other use cases.

Ashby's Data Team

The data team at Ashby is centralized and managed by Kevin. Though they are responsible for internal analyses, reporting, and dashboards, the responsibilities also extend to novel areas like producing global benchmarks and reports on trends in the recruiting industry (many of which make it into the Ashby Talent Trends Report).

The data team serves all teams in the company, with a particular focus on enabling go-to-market teams and scaling data infrastructure as Ashby continues to grow.

Ashby's Data Infrastructure Before Polytomic

As Ashby's customer base spans large public companies and hypergrowth startups, its product datasets are massive and constantly growing. Before Polytomic, Ashby’s data team had an advanced in-house implementation with an orchestration layer to sync data from PostgreSQL to Snowflake. Although this served them well for years, it had three drawbacks:

1. Automatically propagating deletions to Snowflake was difficult.
2. It was not CDC-based.
3. It required engineering time for maintenance.

With Ashby's data volumes rapidly growing, this solution was not sustainable. They needed a dependable platform to sync and ETL data from PostgreSQL to Snowflake that fit these requirements:

1. Highly secure

Ashby houses extremely sensitive candidate and hiring data across its customers. Any solution would have to come with a high-security posture.

2. Highly scalable

Ashby's data volumes are not only large but also growing rapidly, both in aggregate and at local column level with TOAST values. Any solution would not only have to successfully sync today's volumes, but also automatically scale with future growth.

3. Highly reliable

The data in Ashby's Snowflake is queried many times a day, both for internal needs as well as customer ones. Any solution that syncs PostgreSQL to Snowflake would have to be consistently running and reliable, rather than simply doing one-off syncs here and there.

After Polytomic

Ashby surveyed multiple vendors then settled on Polytomic, as Polytomic took care of all the above requirements and more.

One ETL vendor spent many weeks trying and failing to establish a secure connection to our PostgreSQL database that met our requirements. Polytomic, within a week, successfully established the connection and synced our uniquely-challenging data too.

The Python scripts have been replaced by a reliable CDC stream from PostgreSQL to Snowflake. Nothing needs babysitting; Polytomic runs in the background and Ashby's operations confidently execute on current data in Snowflake.

Polytomic's Customer Support

In the ongoing relationship with Ashby, Polytomic's customer support plays a central role. A quote from Kevin sums it up:

Despite the size of their customer base, I’m amazed at how the Polytomic team has scaled customer support. Chat responses within five seconds are not uncommon; feature and integration requests are promptly attended to. On integration requests in particular, it's been comforting to know that we have a partner that quickly builds reliable integrations with new systems that come up at Ashby.

Ashby's Future with Polytomic

As Ashby's growth continues and AI needs multiply, Polytomic will continue playing a central role in data movement. And because Polytomic handles Reverse ETL too, the Ashby team feels secure in not having to buy from yet another vendor to incorporate that functionality.

Polytomic is a central character at Ashby. Whether it's internal or customer data, Polytomic will continue being the main hub we rely on to move data where it's needed.