The page tree for a marketing site: intent, owner, and what gets indexed.
A website sitemap template is an information-architecture worksheet, not a robots.txt companion file. This marketing site sitemap template helps you decide which URLs exist, who they are for, and which ones should never be indexed.
It is the map of templates and URLs before anyone designs a page. Home, solutions, use cases, resources, and legal — each with intent and an owner. XML sitemaps come after the IA is honest.
Path and page type (hub, spoke, legal, app).
What query or job the page answers.
PMM, content, product, legal.
Yes, noindex, or canonicalize to another URL.
The internal-link parent. Orphans do not rank.
A 40-URL SaaS site had three “platform” pages competing. The sitemap forced one hub and two children, and noindexed the leftover comparison drafts.
Harbor (workflow SaaS marketing site)
| Field | Filled |
|---|---|
| Hub | /platform — category + proof, indexable |
| Child | /platform/approvals — job page, indexable |
| Child | /platform/audit-log — job page, indexable |
| Kill | /solutions/platform-overview — 301 to /platform |
| Noindex | /lp/* paid clones |
Crawl or export. You cannot plan a tree you have not seen.
Two pages that answer the same query need a winner.
Every URL gets yes, noindex, or redirect. No “decide later.”
Orphan ownership is how thin pages survive.
Do not submit URLs you just agreed to kill.
| Mistake | Fix |
|---|---|
| Confusing this with an XML sitemap | This is IA. XML is a hint file for crawlers. |
| A row per blog post | Blog is a template + rules, not 400 rows. |
| No redirect column | Killed URLs without a 301 become a support ticket. |
Close. Sitemap here is the URL tree. Site architecture adds navigation, templates, and content types. Use the architecture sheet when you are redesigning the system, not one launch.
No. Thank-you pages, filter URLs, and paid clones usually should not.
PMM plus whoever owns SEO. Engineering needs the redirect list before launch day.