Mega menus and internal link juice

On every page?

[anallawalla]

Large corporate sites often have a mega menu that is repeated on every page of the site, so any page that is linked from it has much the same quantity of internal link juice.

In my ideal world, I would like the navigation bar to have fewer links when the user goes deeper into a branch. e.g. if the user has dropped down to the credit card category page, the top level nav bar remains the same, but the other headings lose their drop-down choices.

If the user then goes deeper in the journey, e.g. a specific kind of credit card, then all other top-level choices disappear.

My thinking is preservation of "credit card link juice" and not distracting the user with other offerings. For a few years, the usual resistance to this internally is from marketers who want the user to see "all options all the time" and developers who would rather not have to code such a nav bar.

To add to the fun, what happens when one of the other navigation choices has a lot of juicy third party links and now you are restricting their effect on other products?

What is the current SEO thinking on this (don't worry about the marketing viewpoint)?

[tangor]

Ultimately it all comes down to usability. Setting up a few testing groups will give information on whether reduction works better than full fat.

Personally I like full fat with commonsense. Each page (and the nav) is a potential landing page and I like users to know there's more to the site that the one page.

As to SEO .. internal. link juice .... testing will determine that.

[keyplyr]

Hi anallawalla,

I have no idea about the SEO thinking (or what SEO even is) but from a user point of view I might get confused about where I was or where I came from if pages had different context menus, especially if it was my first visit to the site.

The menu often gives the 1st time visitor a map of how the site is put together, the hierarchy of pages & where things are.

I can see your point about the Credit Card aspect. But IMO the standard has been established. A menu should be universal. Removing items in the menu on some pages is ambiguous.

[anallawalla]

Thanks, tangor and keyplyr.

To add the conversion angle, I'd prefer that the user is not distracted at that last stage by irrelevant offerings "just in case they don't want a specific personal card but might need to get a business loan or travel insurance". I'm offering three levels of navigation, where the second level still shows all the other offerings but only at their category page, i.e. no drop-downs. You can always click the Home link or the level you just came from.

I have a healthy scepticism of user testing (for site design). I get to see that all the time and it is easy to get the "keep it simple, clean and fresh" response and the conclusion is to remove a lot of words from the site. I have seen money terms drop from #1 to page 2. Then I have to get those words reinstated gently.

Using Amazon as an example, when you go to Books, you still have a "Departments" option that drops down to show *some* other product families and *some* flyouts, but not all. I can live with such a compromise. Walmart has the same approach.

Hence my question was about the SEO link juice angle.

[keyplyr]

Since AI factoring has been introduced, there's been indication that link juice in general has been deprecated. How much is debatable.

I still keep existing internal link schema in place for the user, but I haven't included it in newer projects.

[londrum]

i've just removed a date picker on my pages for the link juice reason. most of my pages had a calendar on the sidebar linking to individual date pages, so that was sixty identical links on every pages.

another reason i did it was because MOZ was flagging up some of the pages as duplicates, even though they each have original content on them. when i removed that big block of code from each page it no longer flagged up. whether google pays attention to that i haven't got a clue -- presumably they're clever enough to discount duplicate content in your navigation bar -- but i'm not taking any chances

[anallawalla]

I like to think of link juice in flavours, i.e. topical link juice. Not too concerned with the numeric value (PR), which we can't see anymore, anyway.

[martinibuster]

That is an interesting proposal. It doesn't make sense to have a link to the page one is on. The problem perhaps is that hand coding this introduces the possibility of human error and you know what that means, entire sections of the site can become unreachable and rankings tank. It also turns something easy into something complex.

The benefit of keeping the mega menu the same is that everything is equal (in terms of link juice).

Internal link juice isn't as important as acquiring deep links into a site, but that has nothing to do with user experience.

Sticking to user experience, I dislike clicking a link and being returned to the same page I was already at, something that has happened to me as recently as yesterday.

My objection to this is the issue of complexity. If there were some way to code this without adding additional CPU or code bloat then it would be easier to see this as viable.

[Robert Charlton]

Very rushed comment for now, but I believe that the key issue for SEO is semantic clarity and ability to emphasize or focus important pages, both for the user and the search engine. This argues against global mega menus.

Some have wondered whether the intelligent surfer model and page segmentation have made mega menus no longer problematic. Lots of test searches I do, as much as one can do on live sites where inbound linking is hidden from me, suggest that unfocussed navigation still clouds search and makes deeper pages hard to find.

I'm still a believer in hierarchical nav, silos, and the inverted-L, which can work together with HTML sitemaps detached from the more focused nav structure.

Also, it should be said that, depending on the type of site, hierachical doesn't necessarily mean top down. Hierarchicies in a sense begin from link entry points. Wikipedia is a case in point. Amazon is another.

Though PageRank has been "deprecated" as a ranking factor, I believe that there are other PR-like factors that still are distributed in a branched fashion. User-attention, in whatever way the search engine measures it, is one such factor.

[tangor]

With the uptake of HTML5 and the <nav> tag it is now semantically possible to isolate navigation from content, which in my opinion means mega menuing will not be detrimental for either the user or SEO.... However, there are still billions of pages out there not coded HTML5 .... so it is quite possible that the search engines have underlying methods of determining content and navigation, particularly drop downs and breadcrumbs.

However, Like Robert indicates above, there is commonsense in controlling nav in reasonable sectioning.

[keyplyr]

I've never seen it negatively affect a site.

[Robert Charlton]

With the uptake of HTML5 and the <nav> tag it is now semantically possible to isolate navigation from content...

True, though the HTML5 community hasn't settled on how to use the tag. Many in the community think that no one has defined what it's for, and that it's pointless. And after that part of the spec is made more specific, browser builders and search engines will still need to get on board.

which in my opinion means mega menuing will not be detrimental for either the user or SEO

Search engines can already pretty much identify drop downs and breadcrumbs.

The issue for me would still be that such a large set of links isn't focused... again, giving both the user and the search engines too many choices.

If you, say, de-emphasize the text content that's enclosed in nav tags, you are weakening the effect of that anchor text uniformly.

In classic onpage SEO, though, pre-mega menu, anchor text on a page was considered an important focusing factor, used mainly for emphasis. But if the sectioned off nav text is such a large mass that it requires deemphasis, that change in the algorithm ends up still having some sort of defocusing effect on the page, albeit in a different direction. I hope that makes sense.

A concern in the HTML5 community about the nav tag, in fact, is that its application to a mega menu probably should differ from its application, eg, to breadcrumbs, and its application in headers and footers should differ, etc. I don't want to sidetrack us with an HTML5 discussion... but it needs to be said that while the HTML5 nav tag might eventually section off links, it hasn't yet been determined what the uses might be.

[tangor]

^^ Thumbs up.

Meanwhile, I am coding for nav to be sectioned off, call me an early adopter. :)

Why? Content is King, ... but that's also a side track! OP is inquiring regards LINK JUICE.