A markdown blog is still one of the most practical content setups for a small SvelteKit site.
It works well when you want to keep writing simple, store posts in the repo, and ship a fast static site without adding a database or CMS too early.
Why use markdown at all?
For a personal site or review blog, markdown gives you a few useful properties:
- posts live beside the codebase
- content changes can be versioned with Git
- pages can be prerendered and served statically
- there is very little maintenance overhead
That tradeoff is usually better than introducing a headless CMS before the site actually needs one.
1. Store posts in a dedicated folder
A clean starting point is a src/posts directory.
Each post should contain frontmatter for the metadata and markdown for the body.
---
title: My Post
description: Short summary shown on listing pages.
date: '2026-03-01'
categories:
- dev
published: true
imageUrl: /img/example.jpg
---
## Heading
Post content goes here.The important part is consistency. Once every post follows the same shape, the rest of the site becomes much simpler.
2. Treat frontmatter as page metadata
The frontmatter should not just exist for the sake of the post page.
Use it for:
- homepage cards
- category sections
- SEO title and description
- Open Graph tags
- publish filtering
That is what makes the content reusable across the whole site instead of only rendering inside a single article view.
3. Load posts through a dynamic blog route
A typical route looks like this:
// src/routes/blog/[slug]/+page.ts
export async function load({ params }) {
const post = await import(`../../../posts/${params.slug}.md`);
return {
content: post.default,
meta: post.metadata
};
}This keeps the route simple:
- the slug maps to a markdown file
- the rendered markdown component becomes the page body
- the frontmatter becomes metadata for the template
The corresponding Svelte page can then render the image, title, date, description, and markdown content in a normal article layout.
4. Generate post lists from the same source
Once the post files are structured, listing pages become straightforward.
A simple API route or loader can:
- import all markdown files
- filter unpublished posts
- sort by date
- group by category
That gives you a homepage, review section, development section, and RSS feed without duplicating content in multiple places.
5. Keep the output content-first
One common mistake is building the markdown system correctly but then wrapping it in pages that feel like templates instead of articles.
For a blog, the better approach is:
- clear summaries on listing cards
- readable article pages
- obvious publish dates
- strong internal navigation
- minimal filler UI
The content should feel like the center of the site, not something placed inside a generic product shell.
6. Know when markdown stops being enough
Markdown is a good fit when:
- one person writes the content
- the publishing flow is Git-based
- posts do not need complicated editorial workflows
It becomes less ideal when you need:
- multiple authors editing in a browser
- scheduled publishing
- approval workflows
- rich embedded content managed by non-technical users
At that point, a CMS may make sense. Until then, markdown is usually the simpler and better system.
Final take
If the goal is a fast, maintainable SvelteKit blog, markdown is still a strong default.
It keeps the stack small, makes publishing predictable, and works especially well for review sites, tutorials, and personal development logs.
The important part is not just rendering markdown. The important part is treating each post as structured content with metadata that can power the rest of the site.
