Typeform accelerates dbt migration from Redshift to Snowflake with Datafold

The problem

Typeform, the form builder that helps over 150,000 businesses collect data people enjoy sharing, was in the midst of a critical evolution: modernizing its data infrastructure by migrating from Redshift to Snowflake. The move was part of a broader mission to enable faster experimentation, deeper insights, and more human data experiences at scale. While the strategic intent was clear, the team faced a common challenge: balancing high-impact initiatives across a fast-growing business, including scaling operations and maintaining legacy systems. Rather than diverting focus from these priorities, Typeform chose a solution that preserved velocity across multiple fronts, accelerating the migration without compromising on quality or momentum.

Mike Angelo, VP of Data, saw an opportunity to change course. “The team had a clear vision for what needed to be done, but made a conscious decision to bring in support so they could maintain focus on high-leverage priorities while executing with precision,” he explained. Rather than diverting full-time staff away from high-leverage, revenue-driving initiatives, Mike set out to find a trusted partner that could not only accelerate execution but also instill greater confidence in the end-to-end developer workflow—from staging to production. The goal: unblock the migration, enable faster iteration cycles, and ship large-scale changes with increased precision and speed.

The solution

While many companies default to consultants for data migrations, Mike explored more agile, value-driven alternatives—seeking a solution that could combine speed, quality, and long-term platform benefit. "I've been a data practitioner for 15-20 years, and these engagements always cost more and take longer than you would expect," he noted.

When Mike discovered the Datafold Migration Agent (DMA), he found the value proposition compelling. "Datafold offers more than just a migration. It also offers a full platform of data quality solutions, including automated testing in CI/CD," he explained. When Datafold’s proposal came in at a fraction of price consultants were quoting while also including the additional platform value, it seemed like the obvious choice.

To further mitigate risk and maximize efficiency, Mike brought in Brooklyn Data Company to handle upstream data engineering work and additional translations. "We needed to kill two birds with one stone and bring in a partner to focus on source pipelines upstream from dbt while giving our Data Engineering team more leverage," he explained.

This multi-vendor approach enabled the team to divide work strategically. As Mike noted: “By segmenting our DAG and reducing cross-team dependencies, we created a clear ownership model that allowed each team to execute in parallel and move faster with fewer bottlenecks."

The results

The strategic partnership delivered significant results, with Datafold's proprietary data diffing technology leveraged across the entire migration project while the Datafold Migration Agent successfully delivered a quarter of Typeform's golden model translations. This multi-vendor approach proved effective, with each team focusing on their areas of expertise while using Datafold's platform for consistent validation in terms of both auditing all of the final business logic used to calculate the company’s key metrics, and using AI-powered features for code validating and rewriting.

Key outcomes achieved:

• Accelerated delivery: Datafold Migration Agent handled complex dbt model translations that would have required significant manual effort, enabling the team to reach milestones much faster
• Universal validation: Datafold’s Data Diff provided a common interface across all teams, enabling consistent validation and quality assurance throughout the migration
• Improved collaboration: Cross-database diffing enabled seamless coordination across globally distributed teams and time zones
• Risk mitigation: Automated Data Diff validation provided confidence that switching to Snowflake wouldn't introduce data drift or quality issues

The team reached their first major milestone—completing all canonical transformations and dbt models—positioning them to go live on Snowflake and eliminate Redshift dependencies by year-end. Beyond the migration, the team has adopted Datafold's CI/CD capabilities, with Staff Analytics Engineer Seth Goldman noting: "We've built up a high degree of confidence running the platform alongside our existing systems.”

The multi-vendor strategy proved successful, with Datafold serving as both a translation partner and the validation backbone for the entire project. As Mike reflected: "I've been in plenty of migrations where we were manually running queries and tracking outputs in a Google sheet. That adds undue stress and time to the project. We wouldn't have gotten to where we needed to go this quickly, or with this level of confidence, without Datafold."

With a faster, more flexible data foundation in place, Typeform is now better equipped to iterate quickly, scale confidently, and deliver the thoughtful, human-centered experiences its customers love. The engagement showcased how Datafold’s modern, AI-powered migration agent can successfully accelerate complex data warehouse migrations and strategic coordination across multiple teams, including systems integrators.