Docusign’s path to 40% cost savings and 60% increased productivity

At Docusign, the goals of the data team are threefold: help leaders make informed decisions, get ahead of market trends and competition, and enable the business to achieve strategic goals.

But none of this is possible without trusted data.

“Every day, I start with one question: have I set things up so my customers can trust the data?” says Bishal Gupta, Analytics Engineering Leader at DocuSign. “If the answer is no, then I still have work to do.”

The Docusign data team knew their legacy infrastructure couldn’t keep up with their data needs, so they decided to modernize their data stack using dbt. In this blog post, we’ll take a look at how that shift has transformed the business, increased cost savings—and even recovered $3 million in lost revenue.

From legacy stored procedures to modular dbt models

It’s a scenario you likely know all too well: complex architecture, legacy stored procedures, and siloed SQL logic.

That’s where Docusign started. These procedures, while functional, were difficult to trace, hard to debug, and introduced inconsistencies across reports and dashboards. This led to a tangled data architecture that stifled scalability and slowed down insights.

“As the business grows, the data grows,” says Gupta. “dbt gave us the agility to adapt, scale, and stay ahead as the business evolved.”

By migrating these procedures to modular, reusable dbt models, Docusign could easily trace and address problems in the data. The result was an organized and flexible data model that enabled faster issue resolution and a streamlined experience for the data team.

“Right off the bat, adopting dbt’s modular approach increased our team’s productivity by 60%,” says Gupta. “Meanwhile, our cost savings improved by 40%.”

Gupta explains that these cost savings came from three factors: first, dbt increased developer productivity by enabling faster model creation. Second, it reduced time spent troubleshooting and addressing data-quality issues. Lastly, dbt's integration with Snowflake optimized data materialization, cutting down on storage and processing costs.

Building a modern data stack by migrating to dbt Cloud

Impactful as the move has been, it was just the first step of adopting a modular, scalable approach to data.

To bring software engineering rigor and a true Analytics Development Lifecycle (ADLC) to their workflows, the data team migrated dbt Core models to dbt Cloud. This transition generated numerous cascading benefits like:

• Improved collaboration. With Git integration and CI/CD pipelines built into dbt Cloud, collaboration across teams became much smoother. Developers can easily manage code versions more effectively and automate deployments with confidence, just like a modern software engineering team.
• Centralized data models. Docusign uses dbt Mesh to create standardized shared models across different teams. Now each team accesses and uses the same trusted base models—ensuring consistency and reducing duplication.
• Standardized macros. By centralizing business logic in standardized sources and reusable macros, Docusign ensures that the same logic is applied consistently throughout the data pipeline. This has helped reduce errors and rework that previously occurred when multiple teams implemented their own versions of the same transformations.

“Stakeholders want as many metrics as possible, as fast as possible—and they want it yesterday,” says Gupta. “Because of dbt, we don’t cut corners to get our stakeholders the accurate metrics they need.”

Deep dive: modularity, reusability, and testability

To illustrate dbt’s impact, let’s walk through how the Docusign data team solved a costly product issue.

When a Docusign customer signs an agreement on the Docusign platform, they often need to submit a payment. DocuSign simplifies this process by allowing them to make payments natively on the platform.

But sometimes these payments wouldn’t go through. This was a huge problem: it created a poor user experience, led to customer churn, and resulted in an estimated $3 million in lost revenue annually.

To tackle this issue, the data, finance, and payments teams decided to work together. Using dbt, the data team:

• Refactored legacy logic into reusable, testable models
• Standardized payment-data pipelines across systems
• Built lineage-aware, production-ready datasets
• Implemented unit and data testing to ensure integrity at every layer

This collaborative, dbt-driven approach helped identify the root cause—and gave teams the tools and confidence to fix the issue quickly and prevent it from happening again.

“Before, we weren’t following the DRY principle of ‘Don’t Repeat Yourself,’” says Guptal. “The result was a weekly headache involving multiple versions of the same data, where the numbers didn’t match.

“So we put the DRY principle in action: we put our code into a macro, which holds all common operations and logics,” Gupta continues. “By using dbt macros and centralized models, we eliminated duplicative logic and reduced errors. Most importantly, we increased productivity and trust.”

The results were striking: the data team improved reporting while also driving significant business impact. The insights they uncovered helped reduce passive churn, identified $3 million in lost revenue, and enabled the business to make better financial decisions.

Lessons for data teams

Docusign’s journey from tangled, complex architecture to a governed, scalable dbt ecosystem illustrates what’s possible when teams align technical vision with business outcomes.

Whether you’re modernizing a legacy stack or migrating to dbt Cloud for the first time, Gupta recommends the following:

• Start small with a proof-of-concept. Prioritize a high-impact use case and use metrics to validate.
• Use the transition to reevaluate business logic. Think through how it will help you scale for new use cases.
• Prioritize modularity, documentation, and test coverage. Focus on the ways you can improve coding efficiency, flexibility, and team productivity.
• Invest in training. Tools like the dbt Fundamentals Certification can help your team quickly learn the foundational steps of transforming data in dbt Cloud.
  

Watch Docusign's Coalesce 2024 session here

If you’re looking to scale a modern data platform, investing in modular analytics engineering with dbt pays off. With dbt, you can build a stack that’s not just smarter—but also more trusted, agile, and impactful. Contact us to book a demo, or sign-up for dbt Cloud to connect your data warehouse and start building.