SGEN SEO architecture overview

⏱ ~2 min answer below · full page ≈ 12 min · skim the bold lead-ins to move faster.

In short. SGEN's SEO surface has five layers: per-page meta, canonical URL handling, automatic sitemap generation, schema.org defaults by content type, and a redirect engine that preserves link equity on slug changes. Everything is automatic by default — canonicals stamp themselves, the sitemap rebuilds on every publish, and renaming a slug creates a 301 redirect without any extra step. You author intent (meta copy, robots directive, redirect destination); the platform carries it forward.

On this page: How SEO works in SGEN · Per-page meta · Canonical handling · Sitemap auto-generation · Schema defaults · Redirect engine · Edge cases and operator habits

How SEO works in SGEN

SGEN's SEO story is five surfaces stacked on top of one another. A search engine crawler touches them in a specific order; each surface contributes a different piece of the indexing picture.

This page covers the concept layer — how the surfaces fit together. For step-by-step configuration, open the relevant module documentation.

Related architecture pages:

Content Modeling overview — how content shape affects schema and URL patterns.
Staging and Versioning overview — why staging URLs are noindex by default.
Analytics Data Pipeline overview — where to read search-source traffic.
Performance & Reliability overview — page speed as a ranking factor.
SGEN Glossary — definitions for canonical URL, sitemap, schema.org, 301 redirect.

1. Per-page meta

Every page in SGEN — whether it lives in pages, posts, or a custom collection — has its own meta block. The block carries: title, description, Open Graph title, Open Graph description, Open Graph image, Twitter card type, and a robots directive (index or noindex, follow or nofollow).

When you save a page, the meta block saves with it. When you publish, the meta block becomes part of what crawlers see. There is no separate "SEO settings" panel in a different corner of the admin — meta lives next to content because they are written together.

Fallback behavior: for pages with no meta authored, SGEN falls back to defaults pulled from the page's first heading and first paragraph. The fallback is rarely ideal — any page worth publishing is worth a deliberate meta block.

2. Canonical handling

Canonical URLs prevent the same content from competing with itself in search results. SGEN sets a canonical URL for every public page automatically. The default is the production URL — clean, lowercase, no trailing slash unless the page lives at a directory root.

The canonical rule set:

Production pages canonical to themselves.
Staging pages carry a robots noindex header so they never enter the index. (No canonical needed; the page is invisible to search.)
Duplicate paths — reachable at more than one URL (trailing slash variant, query string) — canonical to the clean production URL.
Syndicated content — if you publish a piece originally published elsewhere, you can override the canonical per-page to point at the original source.

To verify a canonical: view source on the page and look for <link rel="canonical" href="..."> in the head.

3. Sitemap auto-generation

SGEN generates an XML sitemap at /sitemap.xml on every public site. The sitemap rebuilds whenever the set of published pages changes — a new page published, an existing page unpublished, a slug changed.

In the sitemap:

Every published page, post, and custom-collection item with status: published.
lastmod timestamp from the most recent meaningful edit (content or meta).
A priority hint based on page depth in the site tree.
A change-frequency hint based on edit history.

Not in the sitemap:

Pages marked noindex.
Pages in draft, scheduled, or archived states.
Pages in staging environments.
Pages explicitly excluded via the per-page sitemap-exclude toggle.

The sitemap is referenced from /robots.txt automatically, so search engines find it without operator action.

4. Schema defaults

SGEN emits schema.org JSON-LD for every public page based on the page's type. The defaults:

Page type	Default schema emitted
Article or blog post	`Article` with author, publish date, modified date, headline, image
Product page	`Product` with offers, price, availability, brand
Event page	`Event` with start date, location, organizer
FAQ page	`FAQPage` with question/answer pairs pulled from page structure
Local business page	`LocalBusiness` with address, hours, contact
Generic page	`WebPage` with breadcrumb trail and last-modified timestamp

Schema fields populate from content fields. If an article has an author set, the schema's author populates. If a field is empty, the corresponding schema slot is omitted rather than left blank.

Custom schema: the per-page schema-override block accepts raw JSON-LD that merges with (or replaces) the defaults. This is the escape hatch when your content has a shape SGEN does not template.

5. Redirect engine

URLs change — slugs get rewritten, categories get reshuffled, sections get reorganized. The redirect engine preserves link equity across those changes.

Three flavors of redirect:

Automatic redirects. When you rename a page slug or move a page in the site tree, SGEN creates a 301 redirect from the old URL to the new one without operator action. The redirect persists indefinitely.
Manual redirects. For URLs you want to redirect that were never SGEN pages — legacy URLs from a prior setup, vanity URLs, campaign short links — the redirect manager accepts a from-URL and a to-URL with a chosen status code (301 permanent, 302 temporary, 307 temporary preserving method).
Pattern redirects. For bulk redirects that follow a predictable shape — for example, redirecting everything under /old-section/ to /new-section/ — the pattern syntax handles the substitution in one rule.

Redirect chains (A → B → C) are collapsed at serve time: the engine resolves the chain server-side and sends the visitor straight to C in a single hop.

Edge cases that trip operators up

Canonical and noindex together

Noindex pages still appear in the sitemap by default unless you also flip the sitemap-exclude toggle. Noindex tells crawlers not to index; the sitemap is a hint of what exists. If you want a page hidden from both, flip both toggles.

A canonical override pointing at an external site signals that SGEN is hosting a copy of content that lives elsewhere. Search engines will usually credit the original source. Use the override deliberately, not by accident.

Redirect edge cases

Redirecting from a URL that is currently a published page. SGEN treats the current page as the winner — the redirect rule sits dormant. Unpublish or delete the page first for the rule to activate.

Looping redirects (A redirects to B, B redirects to A). The engine detects the loop at save time and refuses to save the second rule.

Redirects across protocols (http to https). SGEN handles HTTP-to-HTTPS upgrades automatically at the edge. No manual rule needed.

Redirects across hosts (oldbrand.com to newbrand.com). Set up the new domain in SGEN, point the old domain at the same SGEN site, and the redirect manager handles the hostname swap on top of any path-level rules.

Pattern redirects with very broad source patterns (for example /*) can capture URLs you did not intend to redirect. Always preview a pattern redirect against a sample of real URLs before saving. The redirect manager shows the preview before commit.

A 302 (temporary) redirect held in place for years gets treated as a 301 by search engines. If you intend the redirect to be permanent, save it as a 301 from the start.

Schema overrides that produce invalid JSON-LD will be ignored by search engines. SGEN passes the JSON-LD through without validating it against schema.org's spec. Use a schema validator before committing a custom override.

Where SEO surfaces meet operator habits

Two habits separate operators who get consistent search results from those who do not.

Authored meta on every published page. The fallback defaults exist for safety, not as a recommended setting. A page with deliberate meta has a measurably higher chance of ranking for its target query than the same page with auto-generated meta. The cost is one minute per page.

Auditing the sitemap monthly. Open /sitemap.xml once a month and scan. Missing pages = check their status. Unexpected pages = check their robots directives and sitemap-exclude toggle.

The platform handles the heavy lifting. These two habits are the operator's contribution.

Diagnosing common issues

A published page is not showing up in search. Check the page's robots directive first. Publishing alone does not guarantee indexing — the page also needs index, follow in its meta block. If the directive is correct, check the page's age (brand-new pages can take a few days). If it has been published for a week and still does not appear, check Search Console's coverage report.

Two URLs are competing for the same search term. Check the canonical on both URLs. One should canonical to the other. If both canonical to themselves, pick a winner, edit the loser's canonical to point at the winner, and submit both to Search Console for re-crawl.

A rich result (article card, product card, FAQ list) is not showing up. Open the page in Google's Rich Results Test. Most missing rich results come from missing required fields — for example, an article without a publish date will not produce an article card.

A redirect points at the wrong destination. Check the redirect manager for a duplicate rule. If two rules cover the same source URL, the first one created usually wins. Delete the duplicate and clear the edge cache.

The sitemap is missing pages you expected. Check each page's status (drafts, scheduled, and archived pages are excluded by design) and the sitemap-exclude toggle. The sitemap regenerates shortly after publish; give it a few minutes.

How the SEO surface stays accurate over time

URLs are forever. That is the operating principle behind the surface. A page published five years ago should still resolve cleanly — at its original URL or via a redirect — without operator intervention.

The implementation choices:

Automatic redirects never expire. A slug change today still redirects years later.
Canonicals are stamped at serve time, not at save time. Change the canonical rule and every page that uses it reflects the change on its next render.
The sitemap regenerates on every publish event, not on a fixed schedule. The latest published page is in the sitemap within minutes.
Schema defaults follow your content shape. Fill in more fields on an article and the schema grows correspondingly without a separate edit step.

The operator sets intent; the platform carries it forward through time.

A note on link equity

Search engines associate value with URLs, not with content alone. When a URL changes, the value has to flow somewhere — into a redirect to a new URL, into a 410 (gone) that says the content is gone for good, or (worst case) into a 404 that loses the value entirely.

SGEN's redirect engine exists precisely to make the worst case rare. Rename a page: automatic redirect. Delete a page intentionally: the page returns a 410 that tells search engines to drop the URL cleanly. Delete accidentally: the URL stops resolving and the value leaks.

A useful habit: before deleting any page that has external links pointing at it, decide whether it should redirect somewhere relevant. Most of the time, a redirect to the most relevant successor captures more value than letting the URL go dark.