Get AI summaries of any video or article — Sign up free
Launching a SaaS with Almost No Money thumbnail

Launching a SaaS with Almost No Money

Simon Høiberg·
5 min read

Based on Simon Høiberg's video on YouTube. If you like this content, support the original creators by watching, liking and subscribing to their content.

TL;DR

Host the public marketing site on GitHub Pages to avoid expensive website hosting tiers and bandwidth surprises.

Briefing

Launching a SaaS on almost no money is less about finding a magic growth hack and more about engineering the cost structure from day one—then using free (or near-free) marketing channels long enough to build trust. The core playbook centers on hosting and tooling choices that avoid “cloud convenience” markups, plus a marketing mix built around content, SEO, and user-driven distribution.

On the operations side, the first non-negotiable is a website. Instead of paying for expensive hosting tiers, the approach favors GitHub Pages: publish a static site from a GitHub repository and serve it via GitHub’s global CDN at no cost, with no bandwidth surprises. For building that site, programmers can use a static site generator like Next.js, while no-code builders can start with a template generator called Shuffle, export the generated site, and host it on GitHub Pages. The transcript emphasizes that exports from Shuffle stay tidy, while exports from tools like Webflow can be messy.

Next comes the dynamic part of SaaS: the app backend. Many founders default to platforms like Vercel, Supabase, or Heroku, but the cost argument is blunt—these services run on AWS underneath and add aggressive margins. A concrete example is given from the portfolio: a product hosted on Supabase cost nearly $160 per month and suffered reliability issues; after migrating to AWS, the monthly cost dropped to about $12. The recommendation is to use AWS directly when building on cloud infrastructure, because it’s cheaper and more predictable than paying third-party layers.

Still, there’s a broader cost-and-control trend: some bootstrappers are moving away from cloud entirely due to unpredictable pricing and limited transparency. The alternative offered is Coolify, an open-source platform for self-hosting many “cloud-like” capabilities. The suggested setup is to rent a machine on Hetzner and install Coolify, creating a fixed-price “own cloud” environment (described as around $50/month) with fewer surprises.

For automation and workflow, the transcript swaps expensive SaaS tools for lower-cost equivalents: Papple for Zapier-like integrations (with a lifetime purchase option) and Baserow as an open-source Airtable alternative that can be self-hosted. Email delivery is handled via AWS SES, which requires approval but is positioned as far more affordable than services like SendGrid or Mailchimp while offering strong deliverability once approved. For content operations, Notion is used as a lightweight CMS: Notion documents are converted to HTML via API for newsletters and blog posts.

Marketing then leans heavily on free distribution. YouTube is presented as the best free reach channel because long-form content builds trust through parasocial relationships—something short-form struggles to replicate because it’s hard to sustain the 40–60 hours of repeated viewing that builds familiarity. Written equivalents (blog posts, Medium, LinkedIn, Reddit) and SEO are treated as parallel engines. A distinctive SEO tactic is “engineering as marketing”: instead of ranking with articles alone, build small free tools that match search intent (e.g., calculators or downloaders), then funnel visitors to the full product.

To accelerate growth without upfront ad spend, the transcript recommends viral loops (powered-by links, embeds, widgets, or links in support emails), affiliate marketing that shares recurring revenue, and—before launch—networking by being helpful to build relationships. The overall message: control costs with smarter infrastructure, then compound attention through content, SEO, and user-driven referrals until the SaaS can stand on its own.

Cornell Notes

The transcript lays out a low-budget SaaS blueprint built around two constraints: keep infrastructure costs tight and grow using mostly free channels. It recommends hosting the marketing site on GitHub Pages (free static hosting) and using AWS directly for dynamic apps to avoid third-party cloud markups, citing a migration example from roughly $160/month on Supabase to about $12/month on AWS. For founders who want to leave the cloud, it highlights Coolify self-hosted on Hetzner as a fixed-price alternative. On growth, it prioritizes long-form content (especially YouTube) for trust-building, pairs it with SEO (including “engineering as marketing” via small free tools), and adds distribution tactics like powered-by links and affiliate programs.

Why does the transcript treat hosting choices as the make-or-break part of “almost no money” SaaS?

Because cloud convenience layers can add large margins and unpredictable bills. The example given is a product that ran on Supabase at nearly $160/month and crashed frequently; after moving to AWS (the underlying infrastructure Supabase uses), the cost dropped to around $12/month. The same logic extends to other managed platforms like Vercel/Heroku-style services: they run on major cloud providers but charge extra for the wrapper.

What’s the recommended setup for a low-cost website, and why?

Host the site on GitHub Pages by turning a GitHub repository into a website served via GitHub’s global CDN. It’s described as free with no surprise fees or bandwidth limits. For building the site, programmers can use Next.js, while no-code users can use Shuffle to create a site in a no-code editor and export it for GitHub Pages hosting.

How does the transcript propose handling dynamic SaaS hosting without paying third-party cloud markups?

Use AWS directly for the backend rather than relying on higher-level services that sit on top of AWS. The transcript argues that AWS is the biggest cloud provider and that using it directly can save substantial money compared with tools that add “aggressive margin” on top of underlying infrastructure costs.

What’s the self-hosted alternative to cloud infrastructure, and what problem does it solve?

Coolify is presented as an open-source set of tools that replicates much of what people buy from cloud platforms, but can be self-hosted. The suggested approach is renting a machine on Hetzner and installing Coolify, aiming for a fixed monthly cost (around $50/month) and fewer pricing surprises than cloud billing.

Which low-cost tools replace common “automation + database + email” categories?

For automation/integrations, Papple is suggested as a Zapier-like alternative with a lifetime purchase option. For databases, Baserow is positioned as an open-source Airtable replacement that can be self-hosted. For email, AWS SES is recommended: it requires approval for use cases but is described as higher deliverability and far cheaper than alternatives like SendGrid or Mailchimp once approved.

What marketing mix is recommended for bootstrapped growth, and what’s the logic behind it?

Use mostly free channels that compound trust and search visibility. YouTube long-form content is emphasized because repeated viewing builds parasocial familiarity; the transcript cites a study framing friendship formation as 40–60 hours, arguing that long-form makes that feasible. SEO is paired with a twist: “engineering as marketing,” where small free tools match search intent (e.g., calculators or downloaders) and then convert visitors to the full SaaS. Distribution is further boosted with viral loops (powered-by links/widgets/embeds and links in support emails) and affiliate marketing that shares recurring revenue.

Review Questions

  1. What cost-saving mechanism does the transcript claim is behind the difference between managed platforms (like Supabase) and using AWS directly?
  2. How does “engineering as marketing” differ from traditional SEO content, and what conversion path does it rely on?
  3. Which self-hosted stack is proposed as an alternative to cloud pricing unpredictability, and what infrastructure provider is suggested for it?

Key Points

  1. 1

    Host the public marketing site on GitHub Pages to avoid expensive website hosting tiers and bandwidth surprises.

  2. 2

    Build the site with Next.js (coding) or Shuffle (no-code), then export and publish to GitHub Pages.

  3. 3

    For dynamic SaaS backends, prefer AWS directly over managed wrappers like Supabase/Vercel/Heroku to reduce added cloud margins.

  4. 4

    If cloud pricing feels unpredictable, self-host “cloud-like” capabilities with Coolify on a fixed-price Hetzner server.

  5. 5

    Replace costly automation and database tools with Papple (integrations) and Baserow (self-hosted database).

  6. 6

    Use AWS SES for email delivery after approval to keep deliverability high and costs low.

  7. 7

    Grow with long-form content for trust, SEO (including small free tools), and distribution loops like powered-by links plus affiliate revenue sharing.

Highlights

GitHub Pages is presented as a fully free way to turn a GitHub repository into a fast website using GitHub’s global CDN—no hidden bandwidth limits.
A real cost comparison is used to justify moving from Supabase to AWS: about $160/month down to roughly $12/month, alongside improved reliability.
Coolify + Hetzner is pitched as a “fixed-price own cloud” alternative to unpredictable cloud billing.
The marketing strategy leans on long-form trust-building (YouTube) and “engineering as marketing,” where free tools rank and convert.
Viral loops are implemented through practical placements: powered-by links in widgets and links embedded in support emails.

Topics

  • GitHub Pages Hosting
  • AWS Cost Control
  • Coolify Self-Hosting
  • Low-Cost Automation
  • Engineering as Marketing
  • Affiliate Growth Loops

Mentioned