The SaaS Tech Stack for Solo Founders (2026): Tools and Costs

Short answer: a solo founder can run a SaaS on free tiers until the first paying customers, then on roughly $80 a month in fixed tools plus payment fees. Our default stack is Cloudflare for hosting, Supabase for Postgres and auth, Paddle (or Stripe) for payments, Resend for email, PostHog for analytics, Sentry for errors and UptimeRobot for uptime. Payments, not software, become your biggest cost and your hardest migration, so choose that layer most carefully.

Every price and limit below comes from the vendor's own pricing or docs page, checked on 7 Nov 2026. Where we give advice rather than facts, it is marked "Our read".

TL;DR#

• One pick per job. The table below gives a pick and an alternative for 14 jobs, from hosting to an AI coding assistant, each with its free-tier limit and first paid price.
• $0 is real before revenue. Cloudflare Workers gives 100,000 requests a day free (Cloudflare Workers pricing), Supabase Free gives a 500 MB database and 50,000 auth MAUs (Supabase pricing) and Resend Free gives 3,000 emails a month (Resend pricing).
• Watch commercial-use terms. Vercel says its "Hobby plan is for personal, non-commercial use" (Vercel pricing). Better Stack's free plan is labelled "Free for personal projects" (Better Stack pricing). UptimeRobot's free plan says "Good for personal and commercial projects" (UptimeRobot pricing).
• Three scenarios. $0 pre-revenue, about $229 a month (tools plus Paddle fees) with 100 customers on a $20 plan, and about $961 a month at roughly $10K MRR. Arithmetic is shown.
• Lock-in lives in payments, auth and data. Hosting, email and analytics are easy to swap. Subscriptions, password hashes and your database are not.
• Skip the rest until customers pay. A support desk, paid analytics, automation plans and staging servers can wait.

Disclosure: ZeroRaised publishes this guide and is a launch site for bootstrapped products. ZeroRaised has no commercial relationship with any vendor named here and earns nothing if you sign up. Facts come from the linked sources as checked on 7 Nov 2026; prices change, so check the vendor page on the day you sign up.

How to read this guide#

This is the hub. It gives one decision per layer and the cost of that decision, and it links out to deeper comparisons where we have already done the work: hosting in Vercel vs Cloudflare Pages, auth in Clerk vs Supabase Auth and merchant-of-record payments in Paddle vs Lemon Squeezy.

The assumption throughout is a typical solo SaaS: a web app with a login, a marketing site, a subscription plan, a few hundred to a few thousand users, and one person doing everything. If you are building a mobile app, a heavy data product or something with GPU workloads, the shape is the same but the hosting and database rows will cost more.

The solo founder SaaS stack at a glance#

Every free limit and price in this table was taken from the linked vendor page on 7 Nov 2026. "First paid" is the cheapest paid plan or the entry tier of the plan most solo founders would move to.

Stripe note: the Stripe figures in this guide are from Stripe's US pricing page. Rates differ by country (the UK page, for example, lists lower rates for standard UK cards), so open the pricing page for your own country before you do the maths.

Boring tech principles for a one-person SaaS#

"Boring technology" is not a vendor, it is a habit: choose tools that are widely used, well documented and easy to leave. For a solo founder that habit matters more than for a team, because you are the on-call engineer, the support desk and the accountant. Our read, in six rules:

1. Use Postgres unless you have a reason not to. Both database picks in the table (Supabase and Neon) are Postgres. If either vendor raises prices or shuts a plan, a pg_dump and a restore gets your data out. Proprietary databases do not give you that exit.
2. Prefer one platform per layer, not one platform for everything. Bundles (database plus auth plus storage, or hosting plus analytics) save setup time, and that is often worth it. But know which pieces are bundled so you can unbundle one later without moving the rest.
3. Pick the framework you already know. The best stack for shipping in four weeks is the one you can debug at midnight without reading docs. The tools in this guide work with most mainstream frameworks.
4. Default to managed services with free tiers, and self-host only what you enjoy running. Self-hosting saves money and costs time. Time is the scarcer resource for most solo founders before revenue.
5. Keep anything that touches money or identity on standard formats. Store your own user table and customer IDs even if a vendor handles the login or billing. That way a migration is a re-link, not a rebuild.
6. Count the cost per customer, not per tool. A $25 database bill is irrelevant if each customer pays $20. A 5% payment fee on every charge is not. Optimise the line that scales with revenue.

Layer by layer: picks, alternatives and lock-in#

1. Hosting and framework#

Pick: Cloudflare Workers and Pages. The free plan includes 100,000 Worker requests a day and 10 milliseconds of CPU time per invocation, and requests to static assets are "free and unlimited" (Cloudflare Workers pricing). Pages gives free accounts 500 builds a month, one build at a time, up to 20,000 files per site and 100 projects per account (Cloudflare Pages limits). Workers Paid has a minimum charge of $5 a month per account and includes 10 million requests a month, then $0.30 per additional million. We found no commercial-use restriction on either page.

Alternative: Vercel. Vercel is the smoothest home for Next.js, and Hobby is generous: 100 GB of fast data transfer, 1M function invocations and 1M CDN requests a month (Vercel pricing). The catch is the licence to use it. Vercel's pricing page says "Our Hobby plan is for personal, non-commercial use", and its fair use guidelines say "All commercial usage of the platform requires either a Pro or Enterprise plan." The guidelines list "Any method of requesting or processing payment from visitors of the site" and "Advertising the sale of a product or service" as commercial. Pro is $20 a month per developer seat with a $20 included credit and 1 TB of fast data transfer. On Hobby, Vercel says "you are limited to the usage caps and cannot purchase additional usage."

Our read: if your SaaS charges money, budget $20 a month for Vercel or use Cloudflare's free and $5 tiers. Our Vercel vs Cloudflare Pages comparison goes deeper on build limits, framework support and edge cases.

If you need long-running servers or background workers: Railway's Hobby plan is $5 a month and includes $5 of usage; its pricing FAQ says that if usage is $5 or less, "you will not be charged extra" (Railway pricing). Usage is billed at $10 per GB of memory per month and $20 per vCPU per month beyond that. Fly.io no longer has a free tier for new organisations ("New organizations don't have a free tier or a monthly free usage allowance"), gives a $1 trial credit for 7 days, and its smallest always-on machine (shared-cpu-1x, 256MB) is $2.19 a month in its cheapest regions (Fly.io pricing).

Lock-in risk: low to medium. Static sites and standard Node apps move easily. Lock-in creeps in through platform-specific features: edge middleware, Workers-only APIs such as KV or Durable Objects, image optimisation and built-in cron. Our read: use them when they save real time, but keep business logic in plain functions you could run on any server.

2. Database#

Pick: Supabase. The free plan gives 2 active projects, a 500 MB database per project, 1 GB of file storage, 5 GB of egress and unlimited API requests. Free projects pause after one week of inactivity (Supabase pricing). Pro starts at $25 a month and includes 8 GB of disk per project, 100 GB of file storage, 250 GB of egress and $10 a month of compute credits, which covers one Micro instance. Paid projects do not pause. Each extra project adds its full compute cost; Supabase's own example is a Pro organisation with two Micro projects costing $35 a month. Compute add-ons run from Micro at $10 a month to Small at $15 and Medium at $60.

Alternative: Neon. Neon's free plan is permanent with no card, and allows 100 projects, 1 GB of Postgres storage per project (20 GB across all projects), 100 CU-hours per project per month and scale to zero after 5 minutes. Free projects are not paused for inactivity, but compute is suspended if a project hits its CU-hours or 5 GB transfer limit until the next billing period (Neon pricing). The first paid plan, Launch, has no monthly minimum: $0.106 per CU-hour and $0.35 per GB-month of storage. Neon's page estimates a typical Launch bill for a small intermittent database at $15 a month.

Our read: pick Supabase if you want auth, storage and a dashboard in one place. Pick Neon if you only want Postgres, like database branching for previews (its free plan allows 10 branches per project), or run several small projects.

Lock-in risk: low for data, medium for features. Both are Postgres, so your tables export cleanly. What does not export cleanly is the glue: Supabase row-level security policies tied to its auth, storage buckets, edge functions and realtime subscriptions. Keep a migration-ready schema in version control from day one.

3. Authentication#

Pick: Supabase Auth, if you are already on Supabase. It is bundled: 50,000 monthly active users on Free and 100,000 on Pro, then $0.00325 per MAU (Supabase pricing). For most solo SaaS products that means auth costs nothing extra for years.

Alternative: Clerk. Clerk counts monthly retained users, meaning a user only counts if they return at least 24 hours after signing up. Its free Hobby plan includes 50,000 retained users per app and up to 3 dashboard seats. Pro is $25 a month, or $20 a month billed annually, with 50,000 retained users included and $0.02 per extra user from 50,001 to 100,000 (Clerk pricing). You pay for polished prebuilt UI components, organisations and a strong dashboard.

Open source option: Better Auth. Better Auth is a TypeScript authentication framework, and its GitHub README says it "is a free and open source project licensed under the MIT License" (Better Auth on GitHub). You run it inside your own app against your own database, so there is no per-user fee. Its homepage also advertises a paid hosted "Infrastructure" dashboard, and a banner on 7 Nov 2026 read "Better Auth is joining Vercel" (Better Auth). We did not verify the dashboard's prices.

The detailed trade-offs are in our Clerk vs Supabase Auth guide.

Lock-in risk: high. Auth is one of the two layers where switching hurts. Users have passwords hashed by the provider, OAuth connections registered against the provider's callback URLs and sessions issued by the provider. Moving means exporting hashes (if the vendor allows it and the hash format is supported by the destination), re-registering OAuth apps, and probably forcing some users to reset passwords. Our read: keep your own users table keyed by your own ID and store the provider's user ID as a column, so a migration changes one column instead of every foreign key.

4. Payments and billing#

This is the most important decision in the stack, because it is the only layer whose cost grows with every dollar you earn.

Pick: Paddle, as merchant of record. Paddle charges "5% + 50¢ per Checkout transaction" on its pay-as-you-go plan, and acts as your merchant of record, which includes "Global tax and regulatory compliance" and "Fraud protection, chargebacks, and buyer support" (Paddle pricing). Paddle's FAQ says it collects the correct rate of tax and submits sales tax returns. Orders under $10 require contacting Paddle for bespoke pricing.

Alternative: Stripe, as payment processor. Stripe has no monthly fee and charges per transaction. Its US pricing page lists "2.9% + 30¢ per successful transaction for domestic cards", with extra fees for international cards and currency conversion (Stripe US pricing). Other countries have their own rates. For subscriptions you add Stripe Billing, listed at 0.7% of billing volume on pay as you go (Stripe Billing pricing). Stripe is cheaper per charge, but you are the seller, so registering for, collecting and filing sales tax and VAT is your job.

What about Lemon Squeezy? Lemon Squeezy also charges "5% + 50¢" and says "Lemon Squeezy is your merchant of record", with a note that "Some payments may be subject to additional fees" (Lemon Squeezy pricing). Its January 2026 update says the team has been building Stripe Managed Payments, Stripe's merchant-of-record product, and plans to help existing users migrate: "Stripe Managed Payments is here, and it's moving fast" (Lemon Squeezy 2026 update). Our read: if you choose Lemon Squeezy today, expect a migration in the future and factor that in. Our Paddle vs Lemon Squeezy comparison has the full breakdown.

The fee maths on a $20 plan: Paddle takes 5% of $20 ($1.00) plus $0.50, so $1.50 per charge, or 7.5%. On Stripe's US pricing, a $20 charge on a domestic card costs 2.9% of $20 ($0.58) plus $0.30, which is $0.88, plus Billing at 0.7% ($0.14), so $1.02 or 5.1%. The gap of roughly 2.4 percentage points ($0.48 a charge) is what you pay a merchant of record to handle tax for you.

Our read: if you sell to individuals or small businesses across many countries, a merchant of record is usually worth the extra few points in year one. Once revenue is meaningful and you have an accountant, compare that fee against the cost of handling tax yourself.

Lock-in risk: very high. Active subscriptions live inside the payment provider. Moving them is not something you fully control: in the worst case every customer has to re-enter payment details, and some will not. Our read: this is the layer to get right first, not the one to "fix later".

5. Transactional email#

Pick: Resend. The free plan sends 3,000 emails a month, with the note "The Free plan is limited to 100 emails per day", on up to 3 custom domains (Resend pricing). Pro is $20 a month for 50,000 emails, with no per-day sending cap and 10 domains, and $0.90 per extra 1,000.

Alternative: Postmark. Postmark's free developer tier is 100 emails a month with no overages. Basic is $15 a month for 10,000 emails, then $1.80 per 1,000 (Postmark pricing). Postmark's FAQ also mentions a bootstrapped discount: a $75 account credit if you contact support.

Watch the 100-a-day cap on Resend Free. A launch day that brings 150 signups, each needing a verification email, can hit it. Upgrade before a launch, not during one.

Lock-in risk: low. Transactional email is an API call or SMTP. The real asset is your domain reputation (SPF, DKIM, DMARC records on your own domain), which moves with you. Keep templates in your codebase rather than in the vendor's template editor.

6. Newsletter and marketing email#

Pick: Kit. Kit's comparison table lists its free plan for up to 10,000 subscribers (Kit pricing). The first paid plan, Creator, is $33 a month at the 1,000-subscriber tier, or $390 billed yearly, which Kit says saves $78 a year. Prices rise with list size.

Alternative: Buttondown. Buttondown's pricing page says it costs "Absolutely nothing for your first 100 subscribers", charges for active subscribers, and lists add-ons such as paid subscriptions and analytics at $9 a month each (Buttondown pricing). Its base paid tiers are set with a subscriber slider that we could not read in a fetched copy, so check the slider for your list size.

Our read: you do not need a newsletter tool on day one. You need a place to collect emails. A free Kit account does that and gives you room to grow.

Lock-in risk: low. Subscriber lists export as CSV. Automations and sequences do not export; you rebuild them.

7. Analytics#

Pick: PostHog. PostHog says every product has a generous free tier and no credit card is required. The free tier resets every month and includes the first 1,000,000 product analytics events, 5,000 session recordings, 1,000,000 feature flag requests, 100,000 error tracking exceptions and 1,500 survey responses, with 1 project and 1 year of data retention (PostHog pricing). The paid plan has a $0 base price and charges only for usage above the free tier.

Alternative: Plausible. Plausible offers a "30-day free trial, no credit card required", then Starter at $9 a month, Growth at $14 and Business at $19 at the up-to-10k monthly pageviews tier, with "2 months free" on yearly billing (Plausible plans, Plausible pricing).

Our read: PostHog inside the app, Plausible on the marketing site if you want cookie-free traffic numbers and are happy to pay $9. Product analytics tells you what signed-in users do; web analytics tells you where visitors came from. A SaaS eventually wants both.

Lock-in risk: low. You lose history when you switch, not your product. Track a handful of named events (signup, activation, upgrade) rather than relying on auto-capture, and re-instrumenting takes an afternoon.

8. Error tracking#

Pick: Sentry. The free Developer plan is for one user and includes 5k errors, 5M spans, 50 session replays, 1 uptime monitor and a 30-day lookback (Sentry pricing). Team is $29 a month monthly or $26 a month on an annual plan, with unlimited users, 50k errors, integrations and 90-day lookback.

Alternative: PostHog error tracking. If you already use PostHog, its free tier includes 100,000 exceptions a month (PostHog pricing). One fewer tool is a real saving for a solo founder.

Lock-in risk: very low. Error trackers are an SDK and a DSN. Switching is an hour.

9. Uptime monitoring#

Pick: UptimeRobot. The free plan includes 50 monitors at a 5-minute interval, no credit card, and is described as "Good for personal and commercial projects" (UptimeRobot pricing). The Solo plan moves to a 60-second interval; its comparison table lists it as starting at €144 a year.

Alternative: Better Stack. Better Stack's free plan includes 10 monitors and heartbeats and 1 status page, plus 100,000 exceptions and 3 GB of logs, but it is labelled "Free for personal projects ($0/month)" (Better Stack pricing). Additional monitors come in packs of 50 at $25 a month, or $21 billed yearly. Our read: read that "personal projects" label as a hint to move to a paid plan once your SaaS is commercial.

For most solo SaaS products UptimeRobot's free plan plus Sentry's error alerts is enough.

Lock-in risk: none. Recreating ten monitors takes ten minutes. Status page URLs are the only thing customers notice; put yours on a subdomain you own.

10. Customer support#

Pick: plain email, then Help Scout Free. Most solo founders should start with a shared support address that forwards to their inbox. When you want a shared inbox, a knowledge base or saved replies, Help Scout's Free plan "includes 5 users, 1 Inbox, and 1 Docs site", with 100 contacts a month, 10 saved replies and no card (Help Scout pricing). Standard is $25 per user per month.

Alternative: Crisp. If you want live chat on your site, Crisp's free plan includes 2 agents, 100 customer profiles, the website chat widget, a shared inbox and a contact form. Its first paid plan, Mini, is $45 a month per workspace with 4 agents and 5,000 customer profiles (Crisp pricing).

Our read: live chat creates the expectation of instant replies, which is hard to keep alone. Email sets a friendlier pace.

Lock-in risk: low. Conversations are history, not infrastructure. Docs and help articles are the part worth exporting, so write them in Markdown somewhere you control.

11. Planning, code and CI#

Pick: GitHub, including Issues. GitHub Free includes unlimited public and private repositories, 2,000 CI/CD minutes a month, and Issues and Projects (GitHub pricing). Team is $4 per user per month with 3,000 Actions minutes. For a solo founder, keeping the code, the issues and the CI in one place is hard to beat.

Alternative: Linear. Linear's free plan has unlimited members, 2 teams and 250 issues. Basic is $10 per user per month, billed yearly (Linear pricing). Linear is faster and nicer to use; 250 issues is the limit you will hit first.

Lock-in risk: low. Issues export. Your git history is portable by design.

12. Domain and DNS#

Pick: Cloudflare Registrar. Cloudflare says "Cloudflare Registrar does not mark up domain prices at all", supports more than 430 TLDs and includes free WHOIS redaction (Cloudflare Registrar, Cloudflare Registrar docs). Its FAQ also says "all domains on Cloudflare Registrar use Cloudflare nameservers", so to use different nameservers you must move the domain to another registrar (Cloudflare Registrar FAQ). Prices depend on the TLD, so we have not quoted one.

Alternative: keep your current registrar. If you already own the domain elsewhere and are happy with its renewal price, there is no urgency to move.

Lock-in risk: low, but high stakes. Domains are portable by rule, but losing control of one takes your app, your email and your login links with it. Turn on two-factor authentication on the registrar account and turn on auto-renew.

13. Automation#

Pick: Zapier, but only once you need it. Zapier Free includes 100 tasks a month, two-step Zaps and 15-minute polling. Professional starts at $19.99 a month billed annually, or $29.99 billed monthly, for 750 tasks; multi-step Zaps are Professional and above (Zapier pricing).

Alternative: n8n. n8n Cloud Starter is €20 a month billed annually for 2.5K workflow executions, and "A standard, self-hosted version of n8n is available on GitHub" (n8n pricing). The code is "fair-code distributed under the Sustainable Use License", not a standard open source licence, so read it before you build a product on top of it (n8n on GitHub).

Our read: most early automations (new signup to Slack, new payment to a spreadsheet) are better written as a webhook handler in your own app. It is free and testable.

Lock-in risk: medium. Zaps and workflows do not export to other tools. Each one is a small rebuild.

14. AI coding assistant#

Pick: Claude Pro with Claude Code. Claude Pro is $20 a month billed monthly, or $17 a month with the annual discount ($200 billed up front), and Claude Code is listed among its features (Claude pricing).

Alternative: GitHub Copilot. Copilot Free includes 2,000 completions a month with limited chat and agent usage; Pro is $10 per user per month (GitHub Copilot plans). Cursor also has a free tier and a Plus plan at $8 a month (Cursor pricing).

We keep AI to one row here on purpose: pick one assistant, learn it well, and do not stack three subscriptions that do the same job.

Lock-in risk: none. The code it writes is yours. Review it like a contractor's pull request.

Three monthly cost scenarios, with the arithmetic#

These are worked examples, not quotes. They use the prices above as checked on 7 Nov 2026, in US dollars except where a vendor only publishes another currency. Domain renewals, taxes and usage overages are excluded.

Scenario A: pre-revenue, $0 a month#

You are building, testing with a handful of users and collecting emails.

Two warnings. Supabase Free projects pause after one week of inactivity, so a quiet beta can go to sleep. And this stack deliberately avoids Vercel Hobby, because once you add a pricing page and a checkout you are inside Vercel's definition of commercial use.

Scenario B: about 100 customers on a $20 plan ($2,000 MRR)#

You have real revenue, so you move off anything that pauses, limits commercial use or caps sending at 100 emails a day.

That is $229 / $2,000 = about 11.5% of MRR, and $150 of it is payment fees. If you used Stripe with Billing instead, the per-charge cost on Stripe's US pricing for a $20 plan works out at $1.02 (see the payments section), so $102 for 100 charges before tax handling, which you would then do yourself.

Scenario C: about $10K MRR (500 customers on a $20 plan)#

You now care about history, team-ready tools and fewer late-night surprises.

That is about 9.6% of $10,000 MRR. Fixed tools are about 2.1% of revenue; payment fees are 7.5%. Our read: at this stage the stack is not where your margin goes. If you want to cut costs, compare your merchant-of-record fee with the cost of handling tax yourself through Stripe and an accountant, and look at annual plans (customers paying yearly means fewer per-transaction fixed fees).

Lock-in and migration risk by layer#

What to skip until you have customers#

Our read: the tools below are not bad, they are early. Each one adds a bill, a login and a thing to maintain before it can earn its keep.

• A paid support desk. Plain email or Help Scout Free covers your first hundred customers.
• Live chat. It promises instant answers you cannot give while building.
• A separate auth vendor if your database already includes auth. Add Clerk when you need its UI, organisations or enterprise SSO, not before.
• Paid analytics beyond a free tier. PostHog's free tier covers most early products. Look at five events, not fifty dashboards.
• Automation subscriptions. Write the webhook.
• Staging environments that cost money. Use preview deployments and a second free database project until a real customer depends on uptime.
• Kubernetes, microservices or multi-region setups. One region, one app, one database. Scale the boring setup vertically first.
• Paid newsletter plans. Kit Free is enough until your list or your need for automations outgrows it.
• Expensive design tools and premium templates. Your first customers care whether it works.

Before you buy anything, check whether it is on a startup perks list. Our free startup perks guide lists credits and discounts that do not require VC backing.

Week-one setup order checklist#

Set things up in the order that protects you if something goes wrong. This order is ours; adjust it to your product.

Day 1: foundations you cannot easily change

1. Buy the domain (or move it) and turn on two-factor authentication and auto-renew at the registrar.
2. Create a GitHub repo, a main branch and a basic CI workflow.
3. Create a password manager entry for every account you open this week.

Day 2: data and identity 4. Create the Supabase (or Neon) project. Write your schema as migrations in the repo, including your own users table. 5. Turn on auth, set the email templates and OAuth providers you actually need (usually email plus Google).

Day 3: deploy 6. Deploy to Cloudflare (or Vercel Pro if you need it). Connect the domain and set up HTTPS. 7. Set environment variables in the host, never in the repo.

Day 4: email that lands 8. Verify your domain with Resend: add the SPF and DKIM records it gives you, plus a DMARC record. 9. Send yourself a signup confirmation and a password reset, and check both arrive outside spam.

Day 5: money 10. Create your Paddle (or Stripe) account and start verification early; it can take time. 11. Build checkout and the webhook that marks a user as paid. Test an upgrade, a failed payment and a cancellation in test mode.

Day 6: seeing problems 12. Add the Sentry SDK and trigger a test error. 13. Add an UptimeRobot monitor on the homepage and on a health-check endpoint. 14. Add PostHog with three events: signed up, activated (your product's first "aha"), upgraded.

Day 7: talking to people 15. Set up a support address and put it in the app footer and in every email. 16. Add a Kit signup form to the marketing site. 17. Write a one-page privacy policy and terms, and link them from the footer and checkout.

When the checklist is done, you are ready to put the product in front of people. Our guide to launching a bootstrapped app covers what comes next.

Real stacks published by bootstrapped founders#

These three examples come from pages the founders or companies published themselves, checked on 7 Nov 2026. They are useful because they show what a stack looks like when real customers depend on it, and how boring those stacks are.

Healthchecks.io: Django, Postgres and Hetzner, run by one person#

Pēteris Caune runs Healthchecks.io, a cron job monitoring service. Its GitHub README describes it as an "Open-source cron job and background task monitoring service, written in Python & Django" (Healthchecks on GitHub). In his 2022 hosting write-up he says "Since 2017, Healthchecks.io runs on dedicated servers at Hetzner" and "The monthly Hetzner bill is €484" (Healthchecks.io hosting setup, 2022). That post describes PostgreSQL with streaming replication to a standby, uWSGI and HAProxy, plus a small Go app for the ping API. In July 2024 he wrote "Yes, Healthchecks.io is still a one-man business" and that it had "652 paying customers, and the monthly recurring revenue is 14043 USD" (Running one-man SaaS, 9 years in).

Our read: a dedicated-server stack is cheaper at scale but needs real ops skill. Note the dates on these numbers; they are not current figures.

Buttondown: a public, itemised stack page#

Buttondown, the newsletter tool run by Justin Duke, publishes a stack page that lists its tools and what each costs. On 7 Nov 2026 it showed "94,038 / yr · 54 tools", with Django, Postgres ("Primary relational database") and Vue ("Frontend framework for the application UI") among them, plus Postmark, Sentry, Cloudflare and Linear (Buttondown stack). Its changelog on that page includes "Replaced Heroku with AWS ($1,700/mo)." Buttondown's about page says "We are funded by our customers", that it is "not funded by investors or venture capitalists" and "We are profitable" (Buttondown about).

Our read: this is the best public picture we know of how a stack's cost grows with a successful bootstrapped product. Email delivery and hosting dominate, as they would for any email product.

Plausible Analytics: Elixir, Postgres and ClickHouse#

Plausible's README lists "Backend: Elixir + Phoenix", "Databases: PostgreSQL (general data), ClickHouse (analytics)" and "Frontend: React + TailwindCSS" (Plausible on GitHub). Its about page says "We are self-funded and profitable, with no outside investors and no plans to sell" and that it "crossed $1 million in annual recurring revenue in 2022" (Plausible about).

Our read: Postgres for application data and a specialised store only where the workload demands it (here, analytics events) is a good template for any solo founder.

Frequently asked questions#

What is the cheapest SaaS tech stack for a solo founder?#

On published prices checked on 7 Nov 2026, Cloudflare Workers and Pages, Supabase Free (database and auth), Stripe or Paddle (no monthly fee, only per-transaction fees), Resend Free, Kit's free plan, PostHog, Sentry Developer, UptimeRobot Free, GitHub Free and Help Scout Free together cost $0 a month. You still pay for a domain, and payment fees once you have revenue.

Can I use Vercel's free plan for a SaaS?#

Not once it makes money. Vercel's pricing page says the Hobby plan is for personal, non-commercial use, and its fair use guidelines say all commercial usage requires Pro or Enterprise. The guidelines list any method of requesting or processing payment from visitors as commercial. Vercel Pro is $20 a month per developer seat with $20 of included usage credit.

Should a solo founder use Stripe or a merchant of record like Paddle?#

Stripe is cheaper per transaction but leaves sales tax and VAT registration and filing to you. Paddle and Lemon Squeezy charge 5% + 50 cents per transaction and act as merchant of record, handling tax and compliance. Our read: if you sell to consumers or small businesses in many countries and do not want to manage tax, start with a merchant of record.

How much does a SaaS tech stack cost per month?#

In our three worked scenarios: $0 pre-revenue (plus a domain), about $79 a month in tools plus about $150 in Paddle fees with 100 customers on a $20 plan, and about $211 a month in tools plus about $750 in Paddle fees at roughly $10K MRR. Payment fees, not software subscriptions, are the biggest line once you have revenue.

What should I not set up until I have paying customers?#

Skip paid support desks, paid analytics beyond a free tier, paid newsletter plans, a separate auth vendor if your database already includes auth, automation subscriptions, staging environments that cost money, and anything involving Kubernetes or microservices. Set up payments, transactional email, error tracking and uptime checks, because those protect revenue.

Ship the stack, then launch it#

ZeroRaised is a launch directory for products built without institutional VC. When your stack is live and your first users are in, submit your product for free. For more guides on building and growing without investors, join the ZeroRaised newsletter.

Sources#

All sources last checked 7 Nov 2026.

• Vercel pricing, last checked 7 Nov 2026
• Vercel fair use guidelines, last checked 7 Nov 2026
• Cloudflare Workers pricing, last checked 7 Nov 2026
• Cloudflare Pages limits, last checked 7 Nov 2026
• Railway pricing, last checked 7 Nov 2026
• Fly.io pricing, last checked 7 Nov 2026
• Supabase pricing, last checked 7 Nov 2026
• Neon pricing, last checked 7 Nov 2026
• Clerk pricing, last checked 7 Nov 2026
• Better Auth and Better Auth on GitHub, last checked 7 Nov 2026
• Paddle pricing, last checked 7 Nov 2026
• Stripe US pricing and Stripe Billing pricing, last checked 7 Nov 2026
• Lemon Squeezy pricing and Lemon Squeezy 2026 update, last checked 7 Nov 2026
• Resend pricing, last checked 7 Nov 2026
• Postmark pricing, last checked 7 Nov 2026
• Kit pricing, last checked 7 Nov 2026
• Buttondown pricing, last checked 7 Nov 2026
• PostHog pricing, last checked 7 Nov 2026
• Plausible subscription plans and Plausible pricing, last checked 7 Nov 2026
• Sentry pricing, last checked 7 Nov 2026
• UptimeRobot pricing, last checked 7 Nov 2026
• Better Stack pricing, last checked 7 Nov 2026
• Help Scout pricing, last checked 7 Nov 2026
• Crisp pricing, last checked 7 Nov 2026
• GitHub pricing, last checked 7 Nov 2026
• Linear pricing, last checked 7 Nov 2026
• Cloudflare Registrar, Registrar docs and Registrar FAQ, last checked 7 Nov 2026
• Zapier pricing, last checked 7 Nov 2026
• n8n pricing and n8n on GitHub, last checked 7 Nov 2026
• Claude pricing, last checked 7 Nov 2026
• GitHub Copilot plans, last checked 7 Nov 2026
• Cursor pricing, last checked 7 Nov 2026
• Healthchecks.io hosting setup, 2022 edition, Running one-man SaaS, 9 years in and Healthchecks on GitHub, last checked 7 Nov 2026
• Buttondown stack and Buttondown about, last checked 7 Nov 2026
• Plausible on GitHub and Plausible about, last checked 7 Nov 2026