Playbooks
The $0 Tech Stack: How to Build and Launch a SaaS Without Spending a Dime
One free pick for each layer of a SaaS: hosting, database, login, email, analytics, error tracking, payments and file storage. With the real limits, the first thing you will outgrow and the licence traps to avoid. Checked in October 2026.
You can launch a real SaaS with a paying customer and a monthly software bill of zero. Plenty of people do. What they know that beginners do not is which free tools are safe for a business, and which ones quietly ban commercial use or run out of room in week two.
This is the stack I would assemble today, one pick per layer. Each pick includes the limit you will hit first and what to do when that happens. Limits come from the tools' own pages in October 2026, and they change, so confirm before you commit.
Layer 1: Hosting
Pick: Cloudflare Pages. 100 projects, 500 builds a month and 20,000 files per site, with no bandwidth cap listed. You will outgrow the build queue first, since free builds run one at a time. If you need a server that stays awake, Northflank's free Sandbox includes 2 services, 1 database and 2 cron jobs with no sleeping. Avoid Vercel Hobby and GitHub Pages for a paid product, because both restrict commercial use.
Layer 2: Database
Pick: Neon. Real Postgres, with 100 projects, 1 GB of storage per project and 100 compute hours a month. It scales to zero after 5 minutes, so the first request after a quiet spell is slow. You will outgrow the 1 GB first. Supabase is the alternative if you want auth and storage in the same place, but free projects pause after a week of inactivity.
Layer 3: Login
Pick: Clerk. The first 50,000 monthly retained users are free, with a custom domain included and no card. Pricing past that point is the common complaint, and it is a problem you want to have.
Layer 4: Email
Pick: Resend for app emails. 3,000 emails a month, capped at 100 a day, with 3 domains. The daily cap is what bites on launch day. For a newsletter, Kit is free up to 10,000 subscribers, but automations need a paid plan.
Layer 5: Analytics
Pick: Umami, plus Microsoft Clarity for recordings. Umami Cloud's free plan covers 100,000 events a month for 1 website. Clarity has no traffic limits and no sampling. Move to PostHog when you want funnels and feature flags, since its free plan has 1 million events a month.
Layer 6: Errors and uptime
Pick: Sentry for errors and Healthchecks.io for scheduled jobs. Sentry's free plan has 5,000 errors a month for one user. Set a sample rate, or one noisy bug will eat the quota. Healthchecks.io has 20 jobs on its free plan. Skip UptimeRobot's free plan for a business, since commercial use is not allowed on it.
Layer 7: Payments
Pick: Stripe if it supports your country, Polar or Paddle if you want tax handled. There is no monthly fee on any of them. Stripe takes 2.9% plus 30 cents, while Polar Starter and Paddle take 5% plus 50 cents, which is 15 percent of a $5 sale.
Layer 8: Files
Pick: Cloudflare R2. 10 GB of storage and 10 million reads a month, with no egress charge. Heavy read traffic can still cost money, so watch the operations count.
The order I would build in
- Week 1: hosting, database, login. This is a working app.
- Week 2: email and error tracking. This is an app you can trust.
- Week 3: analytics and payments. This is a business.
When something on this list gets too small, you have a good problem: the free tier did its job. Upgrade the single layer that hurts, and leave the rest alone.