35:"BlogCategoryTabs" 36:I[901122,["$2","$7","$8","$9","$a","$b"],"$35"] 37:T184d, In March 2021 a designer partner and I started Codewell: a site where new frontend developers practise on real design templates instead of to-do apps. You pick a design, build it in HTML and CSS, post your solution, and compare it with everyone else's. Sixteen templates were free, nine were behind a paid plan. I wrote all the code. My partner did the designs, the templates and the community. We launched on 9 May 2021 and closed the paid plan in July 2023. Here is what the records say, now that I have gone back and read them. ```timeline 2021-03: Building 2021-05-09: Launched 2021-08: Moved to Next.js 2022-03: Peak, 70 paying users 2023-03: Four-week rewrite, dropped 2023-07: Paid plan closed 2024-03: Last backup ``` ![Codewell's home page in July 2023: log in with GitHub, pick a template, build it.](/blog/codewell/home.jpg) ## The numbers - 23,500 people signed up with GitHub in the first year. 62,000 by March 2024, when I took the last backup. - 346 of them paid: 308 on the monthly plan at $8, 38 on the yearly at $72. At the peak, in spring 2022, about 70 subscriptions were active at the same time. - 25 challenges: 16 free, 9 premium. - 1,466 solutions posted by 888 different people, and 245 comments on them. - An 8,400-reader newsletter, which I ran on Substack and later EmailOctopus. - Search sent 56,000 clicks in the last 16 months the site was up, most of them from India and Brazil. The paying users were mostly in the US. ![Google Search Console for codewell.cc, 8 November 2022 to 7 March 2024: 55.9K clicks, 412K impressions, and "codewell" as the top query.](/blog/codewell/search-console.jpg) ```chart title: Sign-ups per month unit: people source: The Codewell database, dump of 9 March 2024 mark: 2021-05 Launch mark: 2023-07 Paid plan closed 2021-05: 2128 2021-06: 1262 2021-07: 1782 2021-08: 2226 2021-09: 1747 2021-10: 1985 2021-11: 2058 2021-12: 1442 2022-01: 2521 2022-02: 1893 2022-03: 2345 2022-04: 1725 2022-05: 1555 2022-06: 1853 2022-07: 1544 2022-08: 1852 2022-09: 1847 2022-10: 1941 2022-11: 2075 2022-12: 2490 2023-01: 2524 2023-02: 1859 2023-03: 1829 2023-04: 1469 2023-05: 1413 2023-06: 1610 2023-07: 2026 2023-08: 1922 2023-09: 1643 2023-10: 1507 2023-11: 1586 2023-12: 1264 2024-01: 1597 2024-02: 1305 ``` ```chart title: Paying subscribers, active at month end unit: people source: The Codewell database, dump of 9 March 2024 mark: 2022-03 Peak, 70 mark: 2023-01 Decline starts, 51 mark: 2023-07 Plan closed 2021-05: 21 2021-06: 22 2021-07: 33 2021-08: 43 2021-09: 55 2021-10: 66 2021-11: 58 2021-12: 59 2022-01: 65 2022-02: 65 2022-03: 70 2022-04: 67 2022-05: 64 2022-06: 65 2022-07: 65 2022-08: 58 2022-09: 59 2022-10: 65 2022-11: 64 2022-12: 60 2023-01: 51 2023-02: 52 2023-03: 51 2023-04: 46 2023-05: 45 2023-06: 45 2023-07: 6 ``` The decline started in January 2023, six months before I closed the plan: 51 active subscribers that month, 45 by June, and the sign-ups kept coming the whole time. So the top of the funnel was big and the paid conversion was thin: about one in 140 of the people who signed up while the paid plan existed ever paid. ==That ratio, more than anything else, is the story.== ## The stack Everything was TypeScript from the first commit. The stack moved once, three months after launch. | | At launch, May 2021 | From August 2021 | | --- | --- | --- | | Frontend | React (Create React App) on Netlify | Next.js with server rendering, on AWS Lambda@Edge, CloudFront and S3 through the Serverless Framework | | API | Strapi (Node.js) on AWS Elastic Beanstalk | same | | Data | MongoDB, GraphQL through Apollo Client, uploads to S3 | same | | Login | GitHub | same | | Billing | FastSpring | same | | Errors, UI | Sentry, Ant Design | same | Billing ran through FastSpring, because Stripe was not available in my country then. A webhook handled activation and deactivation, and cancellation went through the FastSpring API. There was no handling for failed payments. I only noticed that this week, reading the code again. In spring 2023 I started a rewrite as a Turborepo with Next.js 13, tRPC, Prisma on Postgres and NextAuth. It got as far as login, the challenge list and the editor, and then stopped after four weeks. The paid plan closed three months later. ## What I would do differently **Charge for the thing people came for.** People came for the templates, and 888 of them posted solutions. The paid plan gated the Figma files and the premium templates, which is reasonable, but most never needed more than the 16 free ones. A smaller free tier and a yearly-only paid plan would have converted better than $8 a month against free. ![A challenge page: the Figma file locked behind Pro, everything else free, and the community's solutions underneath.](/blog/codewell/challenge.jpg) **Write the webhook for the unhappy path first.** Activation and cancellation are the easy events. Failed renewals are where the money leaks, and I never handled them. **Do not start a rewrite when the business is slowing.** The 2023 rewrite was an engineer's answer to a product problem. Four weeks of tRPC and Prisma did nothing for conversion. **Keep the data.** The only reason I can write this post is a mongodump from March 2024. Most of the numbers above came out of it in an afternoon; the search figures came from a Search Console export. **Tell people yourself.** Everything that brought people to Codewell was free and repeatable: organic search, 55.9K clicks in its last sixteen months alone, building in public on X, and a community that posted 1,466 solutions and kept each other coming back. None of it needs a budget. All of it needs a weekly slot on the calendar, from day one, next to the one for building. ## What it gave me Codewell was the first SaaS product I built, and it taught me the whole flow, not one step of it. Two years of running a paid product end to end: the frontend, the API, the deployment, the billing, the newsletter, the analytics, the support emails, and the decisions in between: what to charge for, when to say no to a rewrite, what to keep when it is over. That is what I build on now. Every product since has started from it, and so has the habit of writing this down in public. 38:T69b,I wanted a newsletter page that looks clean, loads fast, and doesn’t box me into weird styling limits or pricing tiers. I tried the usual platforms. The pattern was always the same: either it looks great but costs too much, or it’s affordable but feels like a template from 2012. So I gave myself a challenge: ship a production-grade newsletter engine in one day. The stack • Next.js for a fast, SEO-friendly site • Resend to send this email • Postgres + Drizzle for the database (type-safe ORM) • Vercel Blob for image uploads I didn’t reinvent the wheel. I assembled a stack that lets one person move fast. What I shipped in 24 hours This isn’t a static page. You’re using real features already: 1) A better writing and reading experience A Tiptap-based editor with markdown shortcuts, code blocks, and drag-and-drop images, so posts look good on web and in your inbox. 2) Actual community, not a monologue • Threaded comments • Magic-link login (no passwords) • Likes on comments 3) Built-in analytics for growth Open/click tracking via Resend webhooks, attribution (where subscribers come from), and churn tracking so I can learn what works and improve. Why bother building this? Owning the platform gives me superpowers: I can tweak the design, add features overnight, and own the data end-to-end. It’s also a reminder of how far the modern dev stack has come: what used to take a team can now be shipped by one person fast, if you keep scope tight. What’s next I’ll use this newsletter as my build log: what I’m shipping, plus notes on product decisions, marketing insights, and AI integration. P.S. Reply to this email. That loop is the point. — Dewi39:T7f7,

I wanted a newsletter page that looks clean, loads fast, and doesn’t box me into weird styling limits or pricing tiers.

I tried the usual platforms. The pattern was always the same:

either it looks great but costs too much, or it’s affordable but feels like a template from 2012.

So I gave myself a challenge: ship a production-grade newsletter engine in one day.

The stack

• Next.js for a fast, SEO-friendly site

• Resend to send this email

• Postgres + Drizzle for the database (type-safe ORM)

• Vercel Blob for image uploads

I didn’t reinvent the wheel. I assembled a stack that lets one person move fast.

What I shipped in 24 hours

This isn’t a static page. You’re using real features already:

1) A better writing and reading experience

A Tiptap-based editor with markdown shortcuts, code blocks, and drag-and-drop images, so posts look good on web and in your inbox.

2) Actual community, not a monologue

• Threaded comments

• Magic-link login (no passwords)

• Likes on comments

3) Built-in analytics for growth

Open/click tracking via Resend webhooks, attribution (where subscribers come from), and churn tracking so I can learn what works and improve.

Why bother building this?

Owning the platform gives me superpowers: I can tweak the design, add features overnight, and own the data end-to-end.

It’s also a reminder of how far the modern dev stack has come: what used to take a team can now be shipped by one person fast, if you keep scope tight.

What’s next

I’ll use this newsletter as my build log: what I’m shipping, plus notes on product decisions, marketing insights, and AI integration.

P.S. Reply to this email. That loop is the point.

— Dewi

2a:["$","div",null,{"className":"relative min-h-screen px-6 pt-28 pb-20","children":["$","div",null,{"className":"mx-auto max-w-screen-md","children":[["$","h1",null,{"className":"text-[36px] font-extrabold tracking-[-0.02em] sm:text-[40px]","children":"Writing"}],["$","p",null,{"className":"mt-1 mb-8 text-[17px] text-muted-foreground","children":"Notes from building things, with the numbers."}],["$","$2c",null,{"fallback":["$","div",null,{"className":"h-64 animate-pulse rounded-2xl bg-muted"}],"children":["$","$L36",null,{"allPosts":[{"slug":"codewell-what-62000-signups-taught-me","title":"My first SaaS: what 62,000 sign-ups and 346 paying users taught me","description":"I co-founded a practice platform for new frontend developers in 2021, built every line of it, and closed the paid plan in 2023. The numbers, the stack, and what I would do differently.","date":"2026-10-12","category":"Building","readTime":"4 min read","featured":true,"cover":"/blog/codewell/home.jpg","stats":[{"value":"62,000","label":"sign-ups"},{"value":"346","label":"paying users"},{"value":"1 in 140","label":"ever paid"},{"value":"2 yrs","label":"the paid plan ran"}],"content":"$37","kind":"post"},{"kind":"issue","issueId":8,"slug":"i-built-my-own-newsletter-platform-in-a-day","title":"I Built My Own Newsletter Platform in a Day","description":"You’re reading the very first email sent from my custom-built newsletter platform. Here’s why I built it, what I shipped in 24 hours, and what I’m building next.","date":"2025-12-29","category":"Newsletter","readTime":"1 min read","featured":false,"stats":[],"content":"$38","html":"$39"}],"postsByCategory":{"Building":["$2a:props:children:props:children:2:props:children:props:allPosts:0"],"Newsletter":["$2a:props:children:props:children:2:props:children:props:allPosts:1"]},"categories":["Building","Newsletter"]}]}]]}]}]