The Myth: You Need a Complex Stack to Run a Real Business
Here's the myth, stated plainly: if you're running a production application - even a small one - you need a managed Postgres instance, a container orchestrator, a CI/CD pipeline with twelve stages, and at least three AWS services you can't quite explain to your accountant.
This is what the industry tells small business owners, freelancers, and early-stage founders every single day. And it's why simple stacks win arguments but lose mindshare - because the people selling complexity have bigger marketing budgets than the people who quietly ship with SQLite and a $6/month VPS.
Why People Believe It (And They're Not Wrong to Be Cautious)
Let's be fair about where this belief comes from. If you Google "best database for production," every top result assumes you're building the next Instagram. The tutorials, the conference talks, the Medium posts - they all model for scale you'll almost certainly never hit. A bakery in San Jose doesn't need the same infrastructure as Stripe.
There's also a real history here. Ten years ago, SQLite had genuine limitations for web applications. Write-ahead logging wasn't the default. Concurrent write handling was rougher. The advice to avoid SQLite in production was correct - in 2014, for high-concurrency web apps specifically.
But the tooling moved. The advice didn't.
Why Simple Stacks Win in Practice

Here's what a simple stack actually looks like for a small business web app: one Linux VPS (Hetzner, DigitalOcean, or Oracle Cloud free tier), SQLite for the database, Caddy or Nginx for the web server, and your application code in whatever language you know best. Total monthly cost: $0 to $12. Deployment: rsync and a systemd restart.
Compare that to a "proper" setup: managed Postgres at $15-50/month, a container registry, Docker builds, a load balancer you don't need, and a monitoring stack that costs more than your revenue. For a website that gets 500 visits a day, this is like hiring a 747 to fly across town.
This is precisely why simple stacks win for the majority of businesses I work with at Autom84You. A custom site on a single server with SQLite handles thousands of daily visitors, costs almost nothing to host, and - this is the part people underestimate - is dramatically easier to debug at 2 AM when something breaks.
Three Businesses Running "Boring" Tech in Production
Pieter Levels and his $2.85M/year empire. The creator of Nomad List, Remote OK, and PhotoAI runs multiple products generating millions in annual revenue. His stack? PHP, jQuery, SQLite. No framework. No ORM. No Kubernetes. He's talked openly about this - SQLite handles his databases, flat files handle the rest. The internet mocked him for years. His bank account didn't notice.
Litestream and fly.io customers. Ben Johnson built Litestream, an open-source tool that continuously replicates SQLite databases to S3. Companies running on Fly.io use this pattern to get a globally-distributed application with a single SQLite file per region. No connection pooling, no database credentials to rotate, no managed database bill. The backup story - the thing people always worry about with SQLite - is now a solved problem for under $1/month in S3 storage.
Small business sites that just work. I built a booking system for a Bay Area dog grooming business last year. The old setup was WordPress with WooCommerce, a managed MySQL database, three plugins for scheduling, and a $47/month hosting plan that still loaded slowly on mobile. The replacement: a custom PHP application, SQLite, and a $6/month VPS. Page loads dropped from 3.2 seconds to under 400 milliseconds. Monthly cost dropped by 85%. And when the owner wants to back up her entire business, she copies one file. You can see the kind of work I mean at autom84you.com/pages/portfolio.php.
Where Simple Stacks Win - and Where They Don't
I'm not going to pretend SQLite is the right answer for everything. If you're building a system where dozens of servers need to write to the same database simultaneously, you want Postgres or MySQL. If you're processing millions of transactions per second with strict ACID requirements across distributed nodes, yes, reach for the big tools.
But here's the question almost nobody asks: is that actually your situation?
For a restaurant website, a contractor portfolio, a local service business with online booking, an internal tool for a 20-person team, a membership site, a small e-commerce store - SQLite is not just adequate. It's often better than the alternative because it removes entire categories of problems. No database server to keep running. No connection limits. No credentials to manage. No network latency between your app and your data.
Why simple stacks win isn't really about technology preference. It's about matching your tools to your actual problem instead of the problem you imagine you'll have in three years.
The Hidden Cost of Complexity
There's a cost to complexity that doesn't show up on any invoice: your time. Every additional service in your stack is another thing that can break, another dashboard to check, another set of docs to read when something goes sideways.
I talk to small business owners every week who are paying $200-400/month for hosting stacks they don't understand, built by agencies that optimized for their own convenience rather than the client's needs. A Kubernetes cluster for a five-page website. A microservices architecture for an app with one developer. This is where why simple stacks win becomes more than a technical argument - it's a financial one.
A custom website from Autom84You starts at $500, and most of what I build runs on exactly the boring stack described in this article. Not because I can't set up complex infrastructure - I've been doing this for over 20 years - but because the right tool for most small businesses is the simplest one that does the job. If your situation genuinely needs Postgres and Docker and a CDN, I'll tell you that. But if it doesn't, I'm not going to sell you complexity you'll be paying to maintain for years.
What to Actually Do
If you're a small business owner evaluating your tech stack - or paying someone else to manage one - here's what I'd suggest:
Ask your developer what database you're running and why. If the answer is "Postgres because that's what we always use," that's not a reason. It might still be the right choice, but it should be a deliberate one.
Count your monthly infrastructure costs. Hosting, database, CDN, monitoring, email service, DNS - add it all up. If you're a local business spending more than $20/month on infrastructure, something might be oversized.
Consider the bus factor. If your developer disappears tomorrow, can someone else understand and maintain your stack? Simple stacks win this test almost every time. A single-server app with SQLite can be understood by any competent developer in an afternoon. A twelve-service architecture with custom Docker networking takes weeks to onboard.
Test your actual load. Install a basic analytics tool and look at your real traffic. Most small business websites get 50-500 visits per day. SQLite can handle tens of thousands of requests per second on modest hardware. You're probably over-provisioned by a factor of 100.
The tech industry has a bias toward complexity because complexity creates dependency. The more intricate your stack, the more you need specialized help to maintain it. That's good for consultants. It's less good for you.
Why simple stacks win comes down to this: they cost less, break less, and the people who run them sleep better. That's not a myth. That's just math.
If you want a second opinion on whether your current setup is right-sized for what you actually need - or if you're starting something new and want it built simply from day one - reach out at nerd@a84y.com. I'll give you an honest answer, even if that answer is "what you have is fine."
Comments
No comments yet. Be the first to share your thoughts!
Leave a Comment