Phoenix Energy completes critical infrastructure migration in six months with Supabase

<Quote img="kris-woods-phoenix-energy.jpg" caption="Kris Woods, CTO, Phoenix Energy"> We needed a system that could handle serious performance and security requirements — without slowing down our developers. Supabase has given us both. </Quote>

Phoenix Energy (formerly Phoenix Capital Group) operates across oil and gas drilling, mineral rights acquisition, and direct investment through corporate bonds. The team rebranded in early 2024 to reflect its evolution into a full-fledged energy company.

Facing MongoDB's SDK and Data API deprecation with only 12 months to migrate, Phoenix Energy's seven-person engineering team rebuilt their entire data infrastructure on Supabase, completing the migration a month ahead of schedule while maintaining zero downtime for their investor-facing applications.

The challenge

Phoenix Energy ran three business-critical applications on MongoDB's Data API, SDKs, and authentication system: an investor portal at invest.phoenixenergy.com, an internal investment admin system, and Ark, a custom-built CRM that replaced Salesforce.

When MongoDB announced the immediate deprecation of the SDKs and Data API in September 2024, the team faced several critical challenges:

• Complete infrastructure replacement across authentication, database, storage, and API endpoints
• Three production applications dependent on the deprecated stack
• Compressed migration window with only 12 months to evaluate, migrate, and validate
• Performance concerns after early testing showed 20-30× slower response times on suggested alternatives
• User experience risk where any friction in the investor portal could cost conversions

<Quote img="kris-woods-phoenix-energy.jpg" caption="Kris Woods, CTO, Phoenix Energy"> We had initially chosen MongoDB because it was all-in-one: auth, storage, database, endpoints. With that going away, we had to rethink our entire backend infrastructure. But it also gave us an opportunity to build something stronger. And that’s when we found Supabase. </Quote>

Throughout the evaluation period—October through December 2024—the team explored MongoDB via Azure, Firebase, and Fauna. Each alternative came with significant drawbacks, from performance issues to unclear answers about long-term stability.

Choosing Supabase

Supabase emerged as the clear choice, offering the same all-in-one appeal that initially drew Phoenix to MongoDB, but backed by a proven Postgres foundation.

<Quote img="kris-woods-phoenix-energy.jpg" caption="Kris Woods, CTO, Phoenix Energy"> Supabase is exactly what once made MongoDB attractive. We had auth again, all in one place. We had database, we had storage. And it's built on open source Postgres, which has been tried and tested for many years. That gave us confidence. </Quote>

Why Supabase

• Proven foundation: Postgres delivered the stability and maturity the team needed for critical workloads.
• All-in-one platform: Auth, database, APIs, and storage under one roof reduced dependencies and operational overhead.
• Approachable learning curve: Documentation and tooling enabled engineers of varying experience levels to onboard quickly.
• Impressive performance: Proof-of-concept tests exceeded MongoDB benchmarks before optimization.
• Active roadmap: Rapid iteration and an engaged community signaled long-term platform health.
• Direct support: Slack channels with Supabase engineers delivered real-time guidance during migration.

The team also appreciated that rising engineers could quickly become productive with Supabase's approachable API design and clear documentation, limiting onboarding friction during a high-pressure timeline.

The approach

With an August 2025 deadline, Phoenix Energy executed a parallel migration across all three applications using a tightly coordinated plan:

• Custom migration tooling
  • Automated data cycling, syncing, testing, and validation between MongoDB and Supabase
  • Continuous integration checks ensured parity across codebases during the rewrite
• Complete query rewrite
  • Translated every NoSQL query into SQL patterns while preserving feature parity
  • Introduced relational models that improved long-term data clarity and maintainability
• Weekend cutover
  • Final data migration ran over a weekend, with the switch flipped Sunday night
  • Only one day of minor issues surfaced during transition, resolved before Monday investor traffic
• Direct team support
  • Supabase engineers provided hands-on guidance via connected Slack channels
  • Best practices around security, indexing, and performance tuning were incorporated in real time

<Quote img="kris-woods-phoenix-energy.jpg" caption="Kris Woods, CTO, Phoenix Energy"> We transitioned three applications in six months. We built migration tooling, rewrote every database query, and flipped the switch over a weekend. When everyone got back to work Monday, nobody really knew any different. The systems just worked. </Quote>

The results

Phoenix Energy completed the migration in August 2024—one month ahead of schedule—with measurable wins for both the business and engineering team.

• Zero user-facing disruption during and after migration across all investor applications
• Improved performance with noticeably faster responses ahead of further optimization work
• Stronger foundation ready for analytics and AI workloads powered by relational data models
• Team confidence to decommission MongoDB instances immediately after cutover
• Six-month execution from project kickoff to production transition with a seven-person team
• Better developer experience through clearer SQL workflows and faster debugging cycles

<Quote img="kris-woods-phoenix-energy.jpg" caption="Kris Woods, CTO, Phoenix Energy"> We've been very happy with how that's gone. Our experience with the Supabase team has been great. Being able to get responses directly through the connected Slack channel made a huge difference. </Quote>

What's next

With a reliable Supabase foundation in place, Phoenix Energy is focused on expanding its data capabilities and responsibly exploring AI:

• Analytics infrastructure: The team plans to use upcoming Supabase ETL features to build an end-to-end analytics stack without introducing new syncing layers.
• AI workloads: Engineers are testing vector databases and MCP servers in isolated Supabase environments, ensuring compliance with SEC regulations before production rollout.

<Quote img="kris-woods-phoenix-energy.jpg" caption="Kris Woods, CTO, Phoenix Energy"> Supabase ETL is going to be really interesting for us. I'd love for everything to live in the same place so the team can focus on building impactful tools. It really does feel like Supabase is just getting started and I am excited about how we can grow with them from here. </Quote>