The Cartographer's First Draft: On Sketching a Sitemap for the Uncharted Site
We often speak of sitemaps as formal, technical declarations. They are the finished maps we hand to a search engine crawler, a neatly drawn guide to every corner of our domain. But what of the site that has no map? The small, hand-built project, the personal archive, or the nascent business venture whose structure exists only in its creator's mind? For these, the prospect of a formal XML sitemap can feel daunting, a task for which one must first become an expert in a new language.
There is a simpler, more human way to begin. Before a single line of code is written, you can draft your sitemap with the most fundamental tool available: your text editor. This is not about creating a file for the crawler. It is about creating a document for yourself. The goal is to externalize the architecture of your site, to see its pathways laid bare in plain text, and in doing so, uncover the pages that might otherwise remain hidden in the fog of good intentions.
Open a new document. At the top, write the root URL of your site. Hit enter twice. Now, indented with a few spaces or a tab, list the main sections. These are your continents. Under each section, indent again and list the specific pages. These are your cities. Don't worry about perfect syntax or exact URLs. Use descriptive names: ‘/about/history’, ‘/essays/on-silence’, ‘/project/red-barn-photos’. The act of typing them out, of creating a visual hierarchy, forces a clarity that mere thinking does not.
You will inevitably find gaps. You’ll write ‘/products/’ and then pause, realizing you have not yet named the individual products, or considered a filtering system. You’ll list your blog categories and discover one that has only a single, lonely post, prompting a decision: build it out or consolidate? This text file becomes a mirror, reflecting the site's true structure—and its absences. It is a conversation with your own creation.
This plain text draft is the crucial first step. It is the cartographer's initial sketch, made before the precise instruments are brought to bear. It helps you identify orphaned pages with no internal links, deep content that requires a clearer path, and sections that are overgrown and need pruning. Once this human-readable map feels coherent, translating it into the formal XML language that crawlers understand becomes a trivial, mechanical task. You are not just building a file for a robot; you are first understanding the lay of your own land.
Notes & further reading
A few pages I came back to while writing this: