RDBMS overkill for tiny non-profit?

dusting

I love the mission of this non-profit I'm interning with-they run urban farms and vocational training for foster youth and disabled young adults. But the internship description had me worried from the start. The founder wrote it using ChatGPT and doesn't really know what she's asking for. She wants me to build a relational database and automate monthly progress reports, but the whole team is five people with zero technical background, and she's essentially running the program solo.

There are around 20 to 30 participants in the summer program. They fill in worksheets on personal goals, emotional well-being, and horticulture therapy. The founder meets with them one-on-one and wants to generate monthly dashboards for guardians. That's the core need.

My gut says a full RDBMS is overkill for that scale. Even SQLite feels unnecessary. What she really needs is something she can maintain after I leave. So I'm thinking: Google Forms feeding into Sheets, maybe with some basic charts and a simple data dictionary. If we want a bit more structure, Airtable could work-still no real database admin required.

I also suggested keeping the personal worksheets private to the students and instead running separate weekly surveys for general program feedback. That preserves privacy while still giving us metrics for public outreach.

The biggest lesson so far: never build something the client can't use six months later. Overengineering kills small org projects. I'm going to map out their whole process on paper first, walk her through a fake example, and only build exactly what that flow demands. Nothing extra.

For anyone who's done similar work for non-technical teams-what tools would you pitch? And how do you get buy-in for something simple when the founder thinks she needs a relational database?

growthgeek

Honestly, I'd push back on the whole RDBMS idea. Small orgs don't need relational purity - they need something that ties inputs to measurable outcomes. Pitch a flat tracking system that logs every action and spits out a simple LTV or ROAS number. A few columns in Airtable with a pivot table will beat a 6-table schema they can't maintain the moment you leave. Walk the founder through a fake report that shows cost per acquired user vs. lifetime value - that's what actually moves the needle. Over-engineering a database for a team that can't even define their own KPIs is just building debt, not value.

metricsmadness

Building for scale when you're not there is the real trick, isn't it? I've seen this pattern play out a hundred times in affiliate site builds - someone lays down a complex tracking system, then six months later the client can't even pull a basic report. Same principle here: 20-30 students isn't enough volume to warrant a relational database unless you're anticipating tenfold growth inside a year. Otherwise you're just adding complexity that dies when you walk away.

A Google Forms → Sheets pipeline with a few dashboard tabs is not glamorous, but it's survivable. The solo founder can maintain it without a manual. Your separation of personal worksheets from program feedback is the smartest piece in the thread - keeps data hygiene clean without bloating the structure.

What I'd add is: before you wire anything up, map the actual decision points the founder needs to act on. Most orgs that size collect noise because they haven't thought about what a "good" month looks like. If you get that clear, the tech becomes trivial.

ctrjunkie

Honestly, for a solo intern project at that level, even SQLite feels like overkill. No point setting up a server or any of that.

I'd map out the relationships upfront, then use Google Forms + Sheets + Charts. That combo gives you a lightweight relational database without the headache. Run your logic past Gen AI to double-check it, then spend the rest of the time on the user guide and training. Iterate from there.

Way easier to maintain and swap out if they ever grow.

coopplayer

Oh honey, seeing that name pop up with a link - classic bot aesthetic. No personality, no context, just a cold hard URL. It's like finding a stock photo in the middle of a mood board. Total trust-killer for any brand trying to build genuine community.