Discussion starter: a technical SEO question, posted with an answer from Niraj Raut to open the thread. If you have dealt with this on a site, reply with what you saw, especially where it differs.
- Technical SEO
Has cutting a 400-link mega menu down to about 60 links ever produced a measurable lift for deeper categories, or did it just hurt discovery?
Disclosure: Niraj Raut, who posted this answer, runs the SEO consultancy linked at the end of it.
1 reply
-
adminAdmin
Public evidence for a lift from this particular change is thin. No controlled public test that I can point to isolates cutting a mega menu from about 400 links to about 60, and a before-and-after comparison cannot separate the menu from the redesign, hub pages or copy changes that ship in the same release. The controlled results that are public point the other way: published split tests have measured gains from adding internal links to category and content pages, not from removing them.
So a sensible default expectation is that the 60 categories you keep may gain a little, and the 340 you drop are likely to lose something unless another strong internal path replaces the menu link. Discovery is the smaller risk, because Google can still find those URLs through XML sitemaps and through parent category pages where they are linked. The larger risks are weaker internal link signals and slower recrawling for pages that lose a link from every page on the site. The outcome turns on three things: whether Google could see the 400 links in the first place, whether the dropped categories carry search demand of their own, and whether you re-route them through hub pages or simply delete them.
What the published evidence actually shows
SearchPilot’s roundup of winning SEO tests (first published December 2021, updated February 2024) includes two internal-linking results. Adding more than 70 category links to a home page footer produced a reported 5% increase in organic traffic, and increasing related-article links on content hub pages from two to four produced a reported 11% increase. These come from a split-testing platform rather than a single before-and-after comparison, which gives them more weight than a case study.
They still need careful reading. They are winners chosen for a roundup, so they show that link changes can produce measurable effects, not how often they do. The post gives few details about the sites involved. And both tests add links, which is the mirror image of your change: by symmetry they suggest the categories losing their menu link are the ones at risk, but they do not measure a removal.
The warnings against trimming navigation are mostly argued from mechanism. A 2016 Practical Ecommerce column on Gap Inc. brands simplifying their navigation argues that removing whole levels of site access affects the ability to pass authority deeper into a site, and advises a worst-case analysis before removing links, but it reports no measured traffic or revenue outcome.
Separate the cut from the re-routing:
When deeper categories gain after a menu cut, check whether the same release added links from parent category pages, breadcrumbs or “popular in this department” blocks. The gain may come from the links that were added, not from the links that were removed.
Two effects bundled into one change
Cutting a sitewide menu does two things at once, and they push in opposite directions.
Concentration: the hoped-for lift
In PageRank-style models, each page divides the value it passes among its links. With 60 menu links instead of 400, each remaining link on every page carries a larger share. The 60 retained categories, and whatever they link to from their main content, stand to gain.
Removal: the risk
The 340 dropped categories lose a link from every page on the site. They move from one click away to two or three, and if the menu was their only internal path, they are left with sitemap discovery alone. Their internal link signals and recrawl frequency are what you are putting at risk.
Google uses links to find pages and as a signal of relevance, but it does not publish how it weights links in repeated navigation against links in a page’s main content. Many practitioners assume sitewide navigation links count for less than contextual ones. If that assumption holds, it shrinks both effects: the concentration gain is smaller than a simple PageRank sum suggests, and so is the loss. Movements that small can sit inside normal ranking noise, which makes this change hard to measure and easy to misattribute.
Check what Google could see before the cut
Before estimating gains or losses, confirm that the 400 links were ever part of what Google processed. Three situations change the answer:
- The mobile menu loads on tap. Google indexes the mobile version of pages. If the mobile drawer fetches its links with JavaScript only when someone opens it, Google’s rendered page may never have contained them, and cutting the desktop menu changes little for search.
- Menu items that are not links. Google’s JavaScript SEO basics state that “Google can only discover your links if they are <a> HTML elements with an href attribute.” Menu entries built as buttons or script handlers were never passing anything as links.
- Two copies of the menu. Some themes render one navigation tree for desktop and another for the mobile drawer, which doubles the menu’s weight in the HTML. Googlebot fetches up to 2MB of an HTML file and ignores the rest. A 400-link menu is unlikely to approach that on its own, but on pages that are already heavy it is worth measuring.
URL Inspection’s rendered HTML for a few templates shows which menu links Google actually received. If most of the 400 are missing, treat the cut as a user experience decision and measure it with navigation analytics.
Which categories are at risk if they leave the menu
Category situation Likely effect of dropping it from the menu What to do Earns organic clicks and has no internal links outside the menu Largest risk: it loses its main internal path and is likely to be recrawled less Link it from its parent category’s main content before cutting, or keep it in the menu Linked from its parent category page with a descriptive anchor, plus breadcrumbs Moderate: one path shrinks, others remain Drop it and watch its crawl frequency in the logs Thin, near-empty, or a filtered view of another category Low risk, and it may sharpen crawl focus Drop it, and question whether it should be indexable at all Little search demand but heavy use by shoppers Small search risk, real revenue risk Decide with navigation analytics and revenue by entry path Never visible to Google (loaded on tap, or not an <a href>)No change for search Treat it as a user experience decision To fill in the first column, crawl the site twice: once normally and once excluding the navigation. Crawlers that classify links by position (navigation, content, footer) make this quick. Categories with no inlinks outside the menu are your orphan candidates. Join that list with 16 months of Search Console clicks and impressions per category, Googlebot requests per category from your logs, and menu click-through and revenue by category from analytics.
How to cut without stranding the deeper categories
- Give every subcategory a crawlable link from its parent category’s main content, with a descriptive anchor, not only a filter button.
- Keep breadcrumbs as real links.
- Add a “popular in this department” block chosen by demand and margin.
- For the long tail, link a plain HTML category directory from the footer, so every category stays within a couple of clicks of any page.
- Keep every indexable category in the XML sitemap. It helps discovery; it does not replace internal links.
- Render one menu in the HTML and style it responsively, instead of shipping separate desktop and mobile copies.
Do the re-routing first and let Google recrawl the parent pages before trimming the menu. Otherwise the dropped categories could spend weeks with no strong path at all.
Why a clean test is hard, and what works instead
SEO split tests change one group of pages and compare it with a similar group left alone. A global menu appears on every page, so you cannot give half the pages a different menu without splitting the whole site’s link graph. Two designs are workable:
- Stage it by department. Trim the menu entries for a few departments and leave comparable departments untouched. Compare the dropped categories with similar categories in the untouched departments, before and after (a difference-in-differences comparison). Trimming anywhere slightly raises the share of every remaining link, including your comparison group’s, so the control is imperfect, but that contamination should be small next to the effect on the dropped pages.
- Before and after with guardrails. If staging is impossible, use a long pre-period baseline, a control set the change cannot touch (such as product pages), and annotations for everything else that shipped.
The primary metric is organic clicks and impressions for the dropped categories against their comparison group. Secondary metrics are Googlebot requests per dropped category, the number of queries each category ranks for, and revenue by category. The logs are the leading indicator: a sharp fall in crawl frequency for dropped categories shows up before rankings move.
Let the logs set the start date. Count from the point where most affected categories, and the pages carrying the new menu, show a Googlebot request since launch. Then allow at least four full weeks, because weekly shopping patterns need to repeat and a small effect needs more data to separate from noise. Avoid launching into a peak trading period, and check Google’s update calendar: core updates completed in March and May 2026, and one landing mid-test can swamp an internal-linking effect.
What result would change the decision
- Dropped categories that kept a parent-page link hold within noise of their comparison group, and their crawl frequency is stable: extend the cut to the next departments.
- Dropped categories lose clicks against the comparison group: restore the ones with demand, or strengthen their hub links, and re-measure before going further.
- Retained categories gain measurably while dropped ones hold: that is the lift the question asks about. Publish it, because the community has very little data on this change.
Hypothetical example:
A homewares retailer has 12 departments and a 400-link menu. Search Console shows that 85 of the 340 categories proposed for removal earn meaningful organic clicks. A crawl that excludes the navigation shows 40 of those 85 have no other internal link, because the department pages list products and reach subcategories only through filter buttons. Cutting the menu as proposed would leave those 40 with sitemap discovery alone. The team adds subcategory link blocks to each department page first, waits until the logs show the department pages recrawled, then trims the menu for three departments, with three similar departments as the comparison group.
Where I would land
If this were my site, I would not cut the menu to chase a ranking lift that the public evidence does not demonstrate. I would cut it if navigation analytics show shoppers are not using most of those 400 links, and I would make it safe for search by giving every category that earns clicks a crawlable path from its parent before the menu changes. Then I would stage the rollout and let crawl logs and category-level clicks decide whether the second stage goes ahead.
Need help with this on your own site?
Niraj Raut works with businesses in Nepal, Australia, the UK, Europe and the US on technical SEO, ecommerce SEO, local SEO and AI search.
Add a reply Cancel reply
Replies are open to approved members. Log in or apply to join.