From a Metabase instance on a production database to a platform 1,000+ customers use every day.
Metabase sat on a production database replica, maintained by engineers and PMs who had other jobs. Numbers were constantly questioned — similar queries produced different results because there was no single source of truth. We built the modern data stack underneath it (Prefect, Airbyte, Snowflake, dbt), then recommended migrating off Metabase entirely. We moved everything to Omni — including customer-facing embedded reports — in three weeks. The team that was maintaining SQL alongside product work went back to building product.
Not simple questions. A success manager prepping for a coaching call asks for a full performance breakdown with areas of opportunity. An ops lead asks what drove a spike in new customers last month. A finance person asks for a P&L breakdown across hundreds of locations. 760 questions last month, from 30+ people across the company. This is all possible because we chose the right tools, built a solid data model, and gave the AI the context it needs to answer complex questions.
Most data consultants scope a project, deliver it, and leave. We run weekly standups. We own their embedded product analytics. We know which metrics are actually used by their Marketing and Ops teams. The engagement started with a straightforward warehouse migration. Two years later, we're the data team for a company of 150+ people.
We've worked this way with 30+ companies over 6 years.
We'll ask you about your data problems, show you what we'd build first, and tell you honestly if we're the right team for it.
We respond faster than most pipelines refresh.