Alnatura has 492,919 internal links and just 1% of pages hold 48.4% of the authority.

Why internal links decide what ranks

Search engines find and weight pages by following links. External links carry authority into a domain; internal links decide where that authority goes once it has arrived. Add up every internal link on a site and you get an internal PageRank distribution — a map of which pages the site itself treats as important. A healthy graph concentrates that signal on the pages a business actually wants to rank.

This is not a new idea. PageRank — the algorithm Google was built on — is at heart a model of how authority flows through links, and while the public ranking system has grown far more complex, the internal principle still holds: a page linked often, from pages that are themselves well-linked, is treated as more important than one that is barely linked at all. What makes internal linking worth auditing is that it is the one part of this system a site owner controls completely. It costs nothing, needs no outreach, and lives entirely inside the CMS.

The decisive detail is where a link sits. A link inside body copy is an editorial decision — someone judged that one page is relevant to another. A link baked into the navigation bar or the footer is a template decision: it appears on every page regardless of context. Both pass equity, but only the editorial link carries a topical signal. When a graph is dominated by templated links, every page ends up voting for the same handful of destinations, and the site loses the ability to steer authority toward the content that should rank.

It is also the part of technical SEO most often left to chance. Teams obsess over page titles and backlinks while the link graph — the thing that quietly decides which pages ever get a fair shot — is whatever the CMS template happened to produce. That is precisely why an audit is worthwhile: the gap between how a site should distribute its authority and how it actually does is usually large, invisible from the front end, and entirely fixable without touching design or copy.

That is the lens we applied to Alnatura. We crawled www.alnatura.de, reconstructed all 492,919 internal links across 4,143 pages, classified each link by its position in the page, computed a weighted internal PageRank, and checked the live HTTP status of every target. What follows is a reading of that data — not an opinion about the brand, and every figure is reproducible from the crawl.

79.3% of the links never leave the template

Figure 1. Internal links by DOM position (n = 492,919). Navigation + header + footer vs. editorial body links.

Of Alnatura's 492,919 internal links, 79.3% are boilerplate — templated into navigation, header or footer and repeated, unchanged, on essentially every page. Only 20.7% (101,862 links) are editorial, placed inside the body of a page. In other words, for roughly every 5 internal links the site publishes, only one expresses an actual relationship between two pieces of content. Navigation alone accounts for 105,799 links (21.5%), the footer for 116,208 (23.6%), and the header for 169,050 (34.3%).

A high boilerplate share is normal for a brand site built on a fixed CMS template — but the degree matters. At 79.3%, the same menu and footer links appear on every one of Alnatura's 4,143 crawled pages, so internal PageRank is distributed almost uniformly to whatever those templates point at, no matter which page a crawler is on. The body content — where Alnatura could connect a product to its guides, campaigns or category hubs — is doing almost none of that work; across the whole site there are only about 24.6 editorial links per page.

This is where control is lost. A site with a substantial editorial layer can deliberately push equity toward priority products or seasonal campaigns by linking to them from relevant articles. With editorial links this thin, Alnatura has no layer large enough to shape the distribution — the template decides everything. And a template cannot know which products matter this quarter, which guide is converting, or which category is strategic; it simply repeats the same links it repeated yesterday.

There is a second, quieter cost. Editorial links are also how crawlers discover and re-discover content and infer topical relationships; a guide that links out to three related products gives the search engine three contextual paths to follow and three reasons to treat those products as related. Boilerplate links cannot do this, because they are identical everywhere and carry no context. A site leaning almost entirely on its menu is, in effect, asking search engines to understand its catalogue from a navigation list alone.

For every dozen internal links Alnatura publishes, only one is an editorial decision. The rest are the template, repeating.

Where Alnatura's authority pools

Because the graph is template-driven, authority flows to whatever the navigation and footer point at — and it pools extremely. Ranked by computed internal PageRank, the top ten pages on www.alnatura.de hold 18.8% of the site's internal authority, and the top 1% of pages hold 48.4%. The single most-linked page is /de-de/, pulled up by 8,088 inbound internal links and carrying 2.43% of all internal PageRank on its own.

The shape of the curve is the story. After the top handful of pages, recognition falls away fast — the rest of the site competes for what is left. This is a winner-take-most market inside a single domain: a few pages absorb most of the internal signal, and everything else is starved. For a page outside that leading group, the gap is not a position to climb so much as a structural disadvantage baked into the template.

What sits at the top is rarely what a business would choose. The table below lists the ten pages Alnatura concentrates its internal authority on. The point of publishing it is not the exact ordering but the pattern: authority is an asset the site is spending, and right now it is spending most of it on /de-de/, /de-de/maerkte/marktfinder/ and /de-de/ueber-uns/mitarbeit/ — pages that accumulate links because they sit in the chrome, not because anyone decided they were the priority.

Read down the list and the same logic repeats: the pages with the most internal authority are the ones the template guarantees a link on from every page, not the ones a strategist would nominate. Authority is finite — every link the template spends on a chrome page is a link it does not spend on a product, a guide or a campaign. Across 4,143 pages, those small, identical, repeated decisions compound into the distribution shown here, and no individual editor ever chose it.

Table 1. The ten pages Alnatura concentrates its internal authority on — measured, not declared.

What outranks the products

Alnatura links to 1,413 product pages, yet together they hold only 23.21% of the site's internal PageRank. Meanwhile the top of the graph is dominated by non-product pages — /de-de/, /de-de/maerkte/marktfinder/ and /de-de/ueber-uns/mitarbeit/. In internal-authority terms, Alnatura's catalogue is outranked by its own service, information and corporate pages.

This is the signature of a footer- and menu-dominated architecture. Pages that appear in the global chrome — contact, about, legal, help, store or campaign pages — accumulate inbound links on every page of the site and therefore accumulate internal PageRank, regardless of whether anyone searches for them. The products, which mostly receive links only from a listing or a category page, never catch up. The result is that Alnatura's strongest internal signal is spent where it delivers little commercial value.

None of this is malicious or even unusual; it is what happens by default when there is no editorial layer to counterbalance the template. But it means the pages that actually drive the business are the ones least supported by the site's own link graph — exactly backwards from what an internal-linking strategy is supposed to achieve.

Alnatura links to 1,413 products — yet they hold just 23.21% of its internal authority. The chrome outranks the catalogue.

Links that lead nowhere new

Alnatura routes 293 internal links through 3xx redirects — a small share, but every redirected hop still bleeds a little equity; pointing links at final URLs recovers it.

Genuinely broken internal links are negligible — only 0 resolve to a true 404. The story on www.alnatura.de is distribution, not breakage: the graph is intact, it is simply pointed at the wrong places.

What Alnatura's site is actually made of

Breaking www.alnatura.de down by its top-level sections shows where the pages — and the authority — actually live. The largest sections are /de-de (4,045 pages, 99.1% PR), /- (94 pages, 0.7% PR), / (2 pages, 0.0% PR). The table below maps each section's page count against its share of internal PageRank; gaps between the two columns reveal sections that are large but under-supported, or small but over-weighted by the template.

On top of that, 19.9% of pages are near-orphans with only one or two inbound internal links — barely reachable, and almost invisible to search engines. Near-orphans are the clearest symptom of a missing editorial layer: when pages are only ever linked from a single listing, anything that drops off that listing — older content, deeper catalogue, seasonal pages — quietly falls out of the internal graph and stops being crawled with any regularity.

Table 2. Top sections by page count vs. their share of internal PageRank.

Why this compounds

For the pages already at the top of Alnatura's graph, this is a moat: every time the template repeats a link, it reinforces that page's standing, the same compounding loop that makes an incumbent hard to dislodge. For everything else — most of the catalogue and most of the content — the takeaway is uncomfortable but clear: they are not competing for a better position, they are competing to be meaningfully linked at all.

The levers are therefore different from classic on-page SEO. It is less about any single page's optimisation and more about the breadth and consistency of internal support: which pages link to a target, from what context, and how deep in the architecture it sits. The same logic increasingly applies to AI-assisted search, which leans on a site's structure and the relationships its content makes explicit; a catalogue that is only reachable through a menu gives an AI very little to reason about.

There is an economic way to read this, too. The authority a site earns from external links is paid for — with content, PR, time or budget. Spending most of it on legal, contact and navigation pages is the SEO equivalent of renting prime shelf space and stocking it with the returns counter. Internal linking is where earned authority is either deployed against commercial goals or quietly wasted, and for Alnatura it is the cheapest, fastest lever available to change which pages win.

The playbook

None of this needs a redesign. The findings map to a short, low-risk playbook, in priority order:

1. Recover the leaks. Repoint the 293 internal links that pass through redirects to their final URLs.

2. Build an editorial layer. Add contextual, in-content links from high-traffic pages to priority products and hubs. Even a few hundred well-placed links would lift the 20.7% editorial share materially and give Alnatura a way to steer equity deliberately.

3. Promote the catalogue. The pages that should rank — products and category hubs — need inbound links from pages that already hold authority, not just from a single listing.

4. Rescue the near-orphans. Surface under-linked pages from relevant hubs and related-content modules so they re-enter the crawlable graph.

5. Rebalance, don't strip. Boilerplate links are unavoidable and useful; the goal is not fewer footer links but more editorial ones, so the ratio shifts and crawlers get meaningful paths to follow.

6. Measure it continuously. Internal-link health drifts every time the CMS publishes. This teardown was produced with LinkJuice, which maps the boilerplate-versus-editorial split, internal PageRank, broken links and orphans from a crawl of any site — so the same analysis can be re-run whenever the site changes.

The graph is intact — it is aimed wrong

Alnatura's site is not broken in the way a status report would flag — most of it returns 200, and the navigation works. The problem is subtler and more consequential: the link graph is overwhelmingly templated, authority pools on a handful of chrome pages, and the content that should rank is left to compete for the remainder. 79.3% boilerplate is not the exception that proves bad SEO here — it is the default the site has settled into.

The encouraging part is that internal linking is the one ranking factor entirely within the brand's control. Every figure in this teardown points to the same low-cost, high-leverage move: add an editorial layer and let the site decide where its authority goes, instead of leaving it to the template.