Click depth: why old posts sink, and how to bring them back up.

Every site has a shape, and most site owners have never seen theirs. Open your home page, and count how many clicks it takes to reach a post you wrote three years ago without using search. On a typical WordPress blog the honest answer is “a lot”, and the number grows by itself every time you publish.

That number is click depth. It is easy to measure, Google has spoken about it more plainly than about most structural questions, and it responds well to a few afternoons of work. It is also surrounded by folklore, so this guide separates what is known from what is repeated.

What click depth is

Click depth is the smallest number of clicks it takes to reach a page from the home page, following ordinary links. The home page is depth 0. Anything linked from it is depth 1. Anything linked from a depth-1 page, and not already reached, is depth 2, and so on. It is always the shortest route: one link from the menu is enough to make a page depth 1, however many other routes are longer.

That convention, home page 0, is what Screaming Frog’s “Crawl Depth” column uses (depth from the start page), and what Sitebulb reports in its Crawled URLs by Depth graph. Two details trip up comparisons between tools: Screaming Frog counts a redirect as a level, so a link through a redirect adds one; and a tool that starts from /blog/ instead of / will report every post one shallower than a tool that starts from the real home page.

A page with no links pointing at it has no click depth at all. It is an orphan page, reachable only through your XML sitemap or from other sites, and finding those is a separate job covered in our guide to finding orphan pages in WordPress.

Click depth is not URL depth

URL depth is the number of folders in an address. /garden/vegetables/tomatoes/ has three; /tomatoes-in-pots/ has one. People often treat the two as the same thing, and they routinely disagree:

• A recipe blog with /%postname%/ permalinks has every post one folder deep, yet a five-year-old post can be thirty clicks from the home page.
• A SaaS help centre with URLs like /docs/billing/invoices/vat/ is four folders deep, but if the docs home links every article, each one is two clicks away.

Google has said which of the two it cares about. Asked on Twitter in 2018 whether the number of slashes mattered, John Mueller replied that Google does not count slashes in a URL, as Search Engine Roundtable reported. Folder structure is a usability and organisation choice; click depth is the number that describes how your links actually behave. How to group posts into folders, categories and clusters is the subject of our guide to silos versus topic clusters.

What Google has said, and the three-click rule

Three statements are worth knowing, all from John Mueller of Google:

• Easy-to-reach pages look more important (2018). In an office-hours hangout, an ecommerce owner asked whether to link store pages from the home page or one level down. Mueller said Google does not count slashes, but that what matters a little is how easy content is to find, and that a page one click from the home page signals that it is probably relevant, as Search Engine Journal reported. Note the size of the claim: “a little bit”, not a ranking factor.
• Too deep is harder to crawl, but flatter is not always better (2021). Asked about a menu with over a thousand links, Mueller said a structure made too deep is harder for Google to crawl, but that a super flat structure is not better than a reasonable pyramid, and he would aim for the pyramid. That is in Search Engine Journal’s write-up of the hangout.
• Every page you care about needs a link (Google’s documentation). Google’s link best practices say every page you care about should have a link from at least one other page on your site, and that Google can only follow links that are <a> elements with an href.

There is also correlation data from crawl logs. At its 2018 user conference, the SEO platform Botify reported that across the billions of pages it analyses, 58.2% of pages fewer than four clicks from the home page were crawled by Google, against 31.5% of pages more than eight clicks away, according to Botify’s own recap. Treat it as indicative: it comes from Botify’s customers, who skew towards very large sites, and it shows correlation, not cause. Deep pages also tend to be the least important ones, which would produce the same pattern.

The three-click rule

You will read that every page must be within three clicks of the home page. It is a web design rule of thumb from the early 2000s, and there is no evidence for it, for users or for search engines. The Nielsen Norman Group calls it an arbitrary rule of thumb that is not backed by data: users happily click more when each click feels like progress. Google has never given a click number. “Shallow beats deep, and important beats unimportant” is as precise as the evidence allows.

How WordPress buries old posts

On a blog, almost all depth comes from one place: the date-ordered archive. WordPress shows 10 posts per listing page by default (Settings → Reading → “Blog pages show at most”, a setting described on WordPress’s Reading settings page), and how deep the older ones end up depends on what your theme puts at the bottom of each listing.

Worked example: 400 posts, 10 per page

Take a personal-finance blog with 400 posts, latest posts on the home page, and no other links to old posts. That is 40 listing pages.

• “Older posts” links only. Many classic themes use plain older/newer links; WordPress’s own the_posts_navigation() prints exactly “Older posts” and “Newer posts”. Page 2 is one click from home, page 3 is two, and page 40 is 39. The posts on page 40 are one more click: the oldest post is 40 clicks deep, and the average post is about 20.
• Numbered pagination, classic theme. WordPress’s the_posts_pagination() shows one page number either side of the current one, plus the first and last pages (its source sets mid_size to 1). That last-page link changes everything at the old end: page 40 is now one click from home. The deepest posts move to the middle of the archive, around page 21, at 21 clicks. Average: about 11.
• Numbered pagination, block theme. The Page Numbers block inside a Query Loop defaults to two numbers either side, so each click jumps further. The deepest posts, again in the middle, are 11 clicks away. Average: about 6 to 7.

Two consequences most guides miss. First, with numbered pagination your oldest posts are not the problem; the posts from the middle years are. Second, a static front page with the blog at /blog/ adds one click to everything, because the blog listing itself is now depth 1. Category archives have exactly the same arithmetic on a smaller scale: a category with 150 posts is 15 pages long.

How to measure click depth

You need a crawler: a tool that starts at your home page and follows links the way a search engine’s crawler would, recording the shortest path to each URL. Three practical options:

Whatever you use, export the results and count pages at each depth. The distribution matters more than the maximum: one ancient tag page at depth 30 is trivia, 200 posts at depth 12 is a structural problem. If you are choosing between a desktop crawler and a WordPress tool, our comparison of Silotree and Screaming Frog covers what each is for.

Measure it twice

Run the crawl once as normal. Then look at depth counting only links in the body of posts, excluding menus, sidebars, footers and archives. Crawlers can do this with link-position filters or by crawling with those areas excluded. The first number tells you whether a crawler can reach the page. The second tells you whether a reader is likely to. A post at depth 2 only because the footer links every category is reachable, but it is not connected to anything; the difference between those two kinds of links is the subject of our guide to contextual versus navigational links.

Sensible targets by site size

Nobody, including Google, has published a correct depth. These are working targets that follow from what Mueller has said (shallow for what matters, a sensible pyramid for the rest) and from what is achievable on real WordPress sites. Adjust them to taste.

For context on the last row: Google’s crawl budget guide is written for sites with roughly a million or more pages, or 10,000 or more that change daily. Most blogs are nowhere near that, so for them crawl budget is rarely the reason to fix depth. The better reasons are that deep posts get fewer readers and look less important within your own site.

Fixes, in order of leverage

Ordered by how many pages each fix moves per hour of work. On most archives the first two do most of the job.

1. Hub pages in the menu

A pillar or hub page that links to every post on a topic, linked from your main menu, puts the whole cluster at depth 2, permanently. Five hubs covering 300 posts is the single biggest depth change most blogs can make, and it improves the site for readers at the same time.

2. Contextual links from shallow, popular posts

Your most-visited posts are usually shallow already (linked from the home page, menus or many other posts). A link from one of them pulls its destination to one click deeper than the source. Pick the deepest posts that still deserve readers and find a sentence in a shallow, related post that already mentions them. This is slow by hand, but it is the fix that also sends readers.

3. Numbered pagination

If your theme only has older/newer links, switching to numbered pagination roughly halves average depth, as the chart above shows. In a block theme, add the Page Numbers block to the Query Pagination block in your blog and archive templates (Appearance → Editor → Templates). In a classic theme it usually means a theme option, a child-theme edit to call the_posts_pagination(), or a pagination plugin. Google’s pagination guidance asks for ordinary <a href> links between pages and suggests linking each page back to the first. Avoid replacing pagination with infinite scroll or a “load more” button unless the listing pages still exist as linked URLs.

4. Category pages with a curated introduction

A category archive is a paginated list, so on its own it has the same problem as the blog. Give each important category a written introduction with links to its best posts, using the category description if your theme displays it. The posts you link from page 1 of the category become shallow; the rest still depend on pagination.

5. The HTML sitemap: a crawler patch, not a fix

A page listing every post, linked from the footer, puts every post at depth 2 overnight. It is the bluntest fix, and Google’s John Mueller is not a fan: in 2022 he said HTML sitemaps should never be needed, comparing them to loading spinners. His point is that needing one means navigation has failed. It makes the numbers look better without connecting anything, so use it, if at all, as a stopgap while you build hubs. Your XML sitemap is a different thing: it helps discovery but adds no links, so it changes nobody’s depth.

6. Related posts

A related-posts block at the end of each post adds a few links per post. Whether it helps depth depends on how it chooses: blocks that favour recent or popular posts keep linking the same shallow posts, while blocks that match by category or tag reach older ones. Check that the links are in the HTML rather than loaded by script, and do not count on it alone.

A worked before and after

A birdwatching site with 620 posts, 10 per page, latest posts on the home page, “Older posts” links only, and nine categories in the menu. No hub pages. A crawl gave this distribution: 215 posts within five clicks, 405 at six or more, the deepest at 38. The median post was somewhere between six and ten clicks from home.

Over four weekends the owner did four things:

1. Wrote six hub pages (garden birds, waders, identification by song, optics, migration, photography) covering 410 posts. Six was too many for the menu, so they went on a “Start here” page that is in the menu, which puts the hubs at depth 2 and their posts at depth 3.
2. Switched the theme to numbered pagination.
3. Added 60 contextual links from the 20 most-visited posts to the deepest posts that still deserved readers.
4. Wrote introductions for the four biggest categories, each linking 8 to 12 key posts.

After a recrawl, 514 posts were within three clicks and the median was three. The six still at six or more were old site-news posts that belong in no hub; the owner left them alone, which is a perfectly good decision. Depth then stopped getting worse with each new post, because every new post now goes into a hub on the day it is published.

What the owner did not do is claim the traffic change on day one. Depth fixes help discovery and signal importance; any effect on search shows over weeks, and some posts will not move at all because they need rewriting, not linking. Compare equal periods before and after, with a few untouched posts as a baseline.

Keeping it shallow as you publish

Depth is not a one-off fix, because pagination pushes every post one slot further back each time you publish. Three habits keep it in place:

• Every new post goes into a hub on publish day, and gets two or three links from related older posts. That keeps it at depth 2 or 3 forever, rather than sinking with the archive.
• Check the depth distribution every quarter. It is one of the checks in our quarterly internal link audit. Look for the posts that have drifted past your target and give them a contextual link.
• When you add a category or tag, ask what links to page 1 of it. A new category in no menu or hub is a new orphan-in-waiting.

These are the same habits that make internal linking work generally; the internal linking strategy pillar has the full weekly routine. A local plumber with 40 pages can do all of this in an afternoon; a SaaS help centre with 2,000 articles needs it as a standing job.

Where Silotree fits

The free internal link checker crawls your published pages with no account and reports click depth alongside orphans, dead ends, broken links and links through redirects, which is the baseline for everything above. The slow fix, contextual links from shallow posts, is what Silotree does: it suggests each link from a sentence you already wrote in a related post, you approve or skip it, and new posts get suggestions when they are published. It writes nothing inside headings, captions or buttons, saves a revision first, and can undo any batch.

Questions people ask