How Ramp Kept Its Data Pipelines Reliable Through 10x Growth

‍

Ramp is an AI-first finance company. Each week, about 1,500 people at Ramp work with Snowflake, either directly or through agents. Ryan Delgado, Head of Data Platform, says Matia is the team Ramp trusts to manage the pipelines into that warehouse.

A Postgres sync that kept failing

Ramp's previous ETL tool couldn't keep up with the throughput Ramp needed. Its most critical sync, from Postgres into Snowflake, failed unpredictably. Each failure meant a full historical restatement of the Postgres database, which sometimes caused outages lasting several days. The problem got worse as Ramp's data grew.

"Matia can sync data a lot faster and then also show more flexibility and resiliency in their sync. That was huge for us."

One tool for 300 integrations

Ramp runs 300 integrations across 50 to 60 external data sources, including Salesforce, Zendesk, and other CRM systems. Matia is the single tool Ramp uses for all of them.

dbt docs that work at thousands of models

Ramp used to host its dbt docs internally. Once Ramp had thousands of dbt models, the docs stopped working. Ramp now loads its dbt docs metadata into Matia's catalog, where the team can see the full lineage of its dbt models. Ryan calls it "lightning fast," with intuitive views.

Keeping pace with 10x growth

Ramp started using Matia two and a half years ago. Since then, Ramp has grown more than 10x, both in how much data it holds and in how much new data arrives each day.

Ryan calls Matia cost-effective, but says the bigger savings come from reliability: the business misses fewer opportunities because the data behind its decisions is there when it's needed.

“When I say literally everyone in the company benefits from the reliability that Matia provides us, that’s not an exaggeration.”

With Matia running the syncs, Ryan's team spends more time building on top of the data and less time maintaining pipelines.