Paddling Ahead of the Curve

DoSomething has a tradition of innovating with new technologies. We engage with new tech practices and leading service providers to squeeze the most out of our technical resources. And, though we operate with a lean team, we amass more data than most non-profits of our size.

Because of that, it’s important for us to focus on being good stewards of data, rather than managing an ever-sprawling architecture. Instead, right-sizing our architecture has helped us focus time and energy on delivering outsize outcomes at a scale and breadth that has historically been reserved for well-funded startups and large companies. To put it simply, our team can now spend more time moving the needle on supporting young people and developing programs that meet their needs.

Building on a longstanding track record as an early adopter of new solutions has also uncovered some additional, unexpected benefits. Working directly with founding teams - the MotherDuck team, in this case - has led to a deep partnership rooted in genuine support of our core initiatives, fundraising, and operations. It fuels us and keeps us excited and invigorated about what’s next.

Simplicity Scales: Efficient, Practical Data Warehousing

Compared to most nonprofits, our internal expectations around data visibility and performance efficiency are unique because of the strategic importance of data in achieving our mission to fuel young people to change the world. Stakeholders across the organization ask for, and use, data to make informed decisions about the type of programming we bring to our platform and whether it’s resonating with our members.

While the volumes of data we handle pale in comparison to the ‘Big Data’ wave of the last decade, they’re still significant. For example, we had ~4 TB of data in our previous platform. Furthermore, before using MotherDuck, queries against some of our larger tables of web analytics data required prohibitive amounts of time to execute compared to the value they generated.

As a result, we evaluated several established data warehousing alternatives. They were untenable. As a mission-driven non-profit, we don’t have the bandwidth to manage overengineered, intimidating, and setup-intensive distributed systems to power our BI dashboards and internal analytics.

Without a surplus of time, we have no room for error, and anything we add to our stack simply needs to work.

We Found a Duck!

Enter MotherDuck, a DuckDB-powered Data Warehouse purpose-built with efficiency in mind. MotherDuck caught our attention early on with their promise of simplicity, speed, and ease of use for teams who aren’t regularly working with petabytes of data.

As we learned during our incredibly warm, human-first onboarding, the 4 TB of data on our previous platform magically compressed to 1 TB of MotherDuck storage and was no problem for their serverless data warehouse to handle.

Gliding Gracefully through Onboarding Waters

Onboarding to MotherDuck was refreshingly straightforward. Unlike onboarding onto a traditional data warehouse, MotherDuck was easy to work with. We began using the product during Beta and saw remarkable speed-ups in performance, workflow improvements, and a ~20X reduction in our data warehouse costs.

While the duck-themed branding initially caught our attention, we realized there was more than meets the eye. MotherDuck’s “this-should-be-easy” approach to common operations relieved tremendous pressure from our engineering team by getting us up and running almost instantaneously to meet our core requirements.

The user experience is so straightforward that even our non-technical team members have organically started to use the product to structure queries against the data themselves. The friendly UI, Column Explorer, and flexible DuckDB SQL have transformed our ability to engage with our data in an eye-opening and refreshingly self-serve fashion.

The Future of Data is Bright 😎

DoSomething’s ethos aligns with the emerging Small Data Movement, and we believe it could serve other non-profit organizations well, too. Looking ahead, we stand at the crest of a MotherDuck and DuckDB-powered paradigm shift in data and analytics, as most organizations do not possess or process data at the scale for which incumbent cloud data warehouses were designed.

It’s clear that the concept of Small Data and its ‘less is more’ approach are poised to have transformative impact. At DoSomething, we’re particularly inspired by the possibilities of transforming BI and self-service analytics with WebAssembly and embedding a DuckDB-powered database directly in the web browser.

These changes reflect a data-driven future where cutting-edge data processing and analytics are more accessible.

Join Us: Let’s DoSomething, Together

DoSomething is energized by the promise of Small Data to democratize data and empower users to move quickly from question to insight. To our nonprofit network and all organizations seeking nimbler ways to interpret data and deliver value, we hope you’ll join us in proactively seeking out the future of technology.

We are proud to continue charting a path as early adopters in this digital age in service of our mission: Fueling young people to change the world.

Learn how we use our new platform to support our work on our website.