| This 71 message thread spans 3 pages: 71 (  2 3 ) > > || |
|The "Mega Menu" Problem and Google Rankings|
Over the past year and a half, I worked with a number of -950 problems. In doing that, I noticed that many of these sites (though not all) had a characteristic that I started to call the Mega Menu. It comes in two forms, but both of them place a lot of anchor text on every page.
Type #1 just has a long laundry list of links down the left hand side (and often across the footer) and Type #2 uses a CSS hover menu so the page LOOKS neater, but the code still holds many, many links - especially if the menu goes into a second or third level.
Why could this create a problem? I have two ideas. First, we know that anchor text is a heavily weighted element in the algorithm. All that opportunity to go over the top with keywords in anchor text cannot be a good thing. And second, Google reps keep repeating the advice not to go beyond 100 links on a page. Even if more links are now spidered, just think of the pile of semantic confusion that this can throw at the relevance calculations.
So I started looking at the client sites that are really cooking - and guess what, they often are between 30 to 40 links on the home page, no more.
On one well ranked site I was called to help with, we improved even further by dropping from about 60 links in the template to only 22.
Using their analytics program we discovered that one of the home page links got over 60% of the clicks, and 10 of them got 99%! The other were there for "SEO" purposes, but the customers apparently could care less. So we dropped a ton of that anchor text from the site template, and saw rankings get even better on the best terms.
On another site, there was a 3-level hover menu that, even before any content area links, offered 130 anchor tags. We re-thought the Information Architecture and moved the site into a classic "inverted L" format, dropping down to about 45 links per page. Again, everything about the site started to perform better, rankings, stickiness, conversions, you name the metric.
VISITORS LIKE IT TOO
Even without SEO concerns, hover menus can be a problem for visitors. For one thing, they cannot see all their options at any one time. For another, it's easy to end up with several links that have the same anchor text but point to different pages. Now that's confusing both for people and for algorithms!
It can be mind-bending work to to generate a good Information Architecture -- and even harder work to choose optimal menu labels. It's much easier to grow a site by just slapping more links onto the menu. However, from what I've seen, slim and trim is the way to go with menus. Google seems to like it a lot, and a visitor's first impression of such a site is "I can deal with this", not "sheesh, where do I start?"
I'm not saying you can't succeed with a Mega Menu, only that it can be more problematic. Especially when you've been programmed with the SEO mantra of "links, keywords, anchor text, links, keywords, anchor text" for many years, it's easy to go wrong without even noticing.
Anyone else have experiences with different sizes of menu? Maybe you have results that run counter to mine, or may they support mine. I know that we each see our own particular slices of the total web and not others.
Whatever the case, I'm interested in how my Mega Menu idea looks across many markets.
[edited by: tedster at 3:18 am (utc) on July 2, 2008]
|Martin Ice Web|
as I recall an interview with somebody from yahoo I remember that they take about 50-100 templates from the internet where they say that are good pages.
The crawled pages are now compared with the 50-100 templates and will now be calculated with on page factors ... for the rankings.
IMO google will do the same?! So there must be at least one template with our MEGA MENUE. And while there are a lot of these pages around this should not be punished by google.
Back to our question. In my niche are many Mega Menues and they do rank very well. 90% have their Menu on left side with more than 100 links per page.
IMO there must be some "hetro" anchor-factor. Because the last google update killed many from the pages with widget red, widget yellow, widget green....
In short the doorways from some of my competitors with only 5-6 links are gone.
|And while there are a lot of these pages around this should not be punished by google. |
I don't think it's punishment, at least not intentional. Instead I think that Mega Menus make the algo's job harder, just as they make the visitor's job harder. And sometimes, it's easier to be "collateral damage" from some algorithm factor that is designed to catch scraper pages or whatever.
I should also make it clear, those redesigned sites I worked with are not small, or even smaller than they were. Some are in the area of several hundred thousand pages.
Interesting that your niche has many successful sites with long menus. That's what I was wondering about. Google's algorithm today does seem quite adaptable to the "local customs" in various markets. thanks for the report.
|Martin Ice Web|
if you reduce the number of links on first level stage pages, then you have to implement more level pages? Is that right? But google now crawls depper pages only if your pagerank is adequate. So this is an never ending story. Reorganize our Menue and have more level pages witch are not crawled ( if you do not have enough pagerank or build up a new side ) or do have a mega menu and will be sorted out as "collateral damage".
I think that a straight menue on left pageside is a very good "fast key" information for the user. If you look at newspaper pages where are no menue I donīt get the clue. No straight line. Only headerlines that go deeper.
Can you give us some more information about the "make the algo`s job harder"?
If every page has essentially the same large collection of anchor text, that blurs the relevance signal of the page. I mentioned this to a Yahoo VP at PubCon, and his eyes rolled back. He said, essentially "spread that idea around. Those menu heavy pages are a major headache for us."
My point is also that we can focus PageRank where it really matters, not spread it so thin that nowhere stands out as a key page on the website. There's a parallel in physical world retail shops (I have decades in that kind of management) where you want to put everything in the front of the store and at eye level. But of course, in the physical world, you just can't do that. So you learn, you do split tests, you see what the best performers are in your inventory.
And that's part of what I like to do with a website. We know there are certain things that render well on the visible page but are not really spider friendly. I think there are parallels with site structure, too. Too big a manu and you can easily make an algorithm ask "where's the structure?"
Our pages can be like our kids - every one of them seems precious and we want the whole world to love all of them -- but it ain't going to happen. We can squeeze so tight trying to force the results we visualize that we crush what is good. To a visitor, a huge assortment of links can come across as almost desperate.
Another analogy I make sometimes is inviting people to your house for a dinner party. You don't tell them to raid the fridge and suit themselves. No, you prepare dishes, you coordinate courses. You create a "menu"!
All I can do is report on my experiences, and yes, I am biased by my own past. But this approach, as I said, has helped at least some successful websites to become more successful. So I feel there's something to it. I'm not just building castles out of sand here, there's some real data, real experience of several years that I'm trying to get deeper into.
And when it comes to PR circulation, all PR does not come from the home page. Great inner pages can attract lots of backlinks and become stars on their own. If you notice those pages, then their links can be great sources of PR circulation, too.
the pages that you redesigned were they very thin on content. I've noticed in my niche that larger menu's do seem to have an effect on pages if the percentage of <p> text compared to the html can have a huge factor on whether the page has value, just wondering whether these sites could have been thin on the ground with actual information.
We in the past have just added content to our pages so in turn doing the same thing but this can be extremely expense and can be obviously scraped.
Good question. These sites were all ecommerce - and the product pages were definitely thin on some of them. But not on all, and they all had some excellent content pages as well.
You're onto something there. Enough text content of the right kind can kind of "ground" the page's unique relevance signals.
I'm working with a new site right now where we're experimenting with articles - some are syndicated only, some are only on the site itself, and some are both. We redesigned the template menu to help bring the article pages more link juice, and we removed template links to many other pages that weren't performing well anyway, pushing them further down the click path.
We've skimmed off 35 links from the template so far, going from 110 to 75 (I still think that's too many, but what are you gonna do, you know?) It's a little early to declare a victory on this one, but early signs are good. In fact, ranking for the Home Page has improved for the two main terms, and one of the new Articles is also starting to get better search traffic on some very feisty search terms.
There was a video I watched about 6 months ago that showed something similar.
They took some e-commerce sites that had templates with large amounts of navigation and somewhat small amounts of content. When they broke the page down, there ended up being more "content" in the navigation (link text, other misc text that showed on every page) than there was in the product description area. Their solution was to change whatever you could to images (thereby reducing site template "content") and to try and get the amount of highlightable text in the "product area" to have more words than the amount of highlightable text in your navigation/site template.
They said rankings increased after this happened and I think it sounds like a similar observation.
In a nutshell, the more things you can do to keep the navigation from "out-bulking" the page's unique content, the better.
Beefing up the content, fewer navigation links and more concise anchor text are part of the solution, but also give thought to how the URLs in your navigation are constructed in the first place.
In my experience (always that qualifier!) a menu full of links like this:
will have a lower threshold for causing problems than a menu full of concise URLs like this:
Leaner, more concise navigation will improve the page loading speed, which is also A Good Thing.
My primary client has around 2500 products, and we have a 'mega menu' of about 200 categories. We don't do it for search engine reasons, but more usability - this is essentially a B2B site (similar to office supplies but for a specific niche, and the products and categories are all over the map) We want people to hone in on the product they want as quickly as possible.
Our SERPS are pretty good - way way way better than any of the competitors, but they certainly could be improved in some areas.
Me, I'd love to have a shorter list. I'm just not sure how to balance that with getting the customers to the page they want without a lot of extra clicks.
Also, this is a catalog company, so while we have a good amount of business coming from Google et al, we also want to make it easy for the folks to who have a catalog in their hands.
One thing I've considered is to change or add navigation by industry or application, rather than by product. But that could upset the long time customers. I keep thinking how ticked off I get because every time I go into my local warehouse club to shop, they've rearranged everything and I can't find what I came to get quickly. And it seems to happen every month.
I'm also somewhat afraid to experiment with a live site that is working pretty well - it could be better, but it could also be much worse.
It's not easy.
I have a question as I have the same type navigation. What if the nav links on the left side number 75 and the total count of links on the page will then number about 90-100 some I have no follows on.
I know all the 100 link thing so this isn't part of the question.
My question is since Tedester you indicated the links you removed were heavy to anchor text and not necessary really for getting around the site in my case mine could and could not be conssidered anchor text. My nav is Manufacture Names or Company name brands.
If you think this may be a problem what suggestions would you have making it a dropdown list and taking out the Company names.
Would this really reduce the number of links per page as most searches I get are product names with the manufacture name.
I have wondered if removing them would benifit the site and or hurt the site any suggestions.
This is a hard call as the links are the way a visitor can find the company and then the product they are searching removing the links and adding a dropdown does this really help?
Dog tedester you opened a can of worms here for me now. Really hard for me to figure how I would set up a good way now to navigate around the site without the links. I need to think this over.
Ok I have thought it over a few minutes and will remove the links from the Home page as I have a Link to a Catalog Page that has the companies listed.
Tedester come to think on it I have tried everthing and I mean everthing I can think of except removing these links from the navigation trying to get this filter lifted off my site.
I will change the rest of the product pages to refelct the change and wait and see.
Great topic tedster!
|However, from what I've seen, slim and trim is the way to go with menus. |
From my perspective, it has always been that way. Many jumped on that Mega Menu bandwagon and it is clearly a challenge for any website that has a larger taxonomy.
I've always looked at it from a "segmentation" viewpoint. Larger sites need special treatment when it comes to information architecture. "Mother" now becomes a "Super Mom" with a host of Children, Grandchildren, Great Grandchildren, etc. not to mention all the "Relatives" that are now actively involved with the family.
The Children, the top level categories, all need to be treated separately, they are their own entity. Instead of feeding a Mega Menu on all pages, you can feed category specific menus at each of the primary category levels. There is no need for the top level category pages to be bulked up with links from all the other categories. No, when someone clicks on one of those primary category links, they are taken to a page that is the "Home Page" for that category. Now the Grandchildren come into play. And from there, the Great Grandchildren and all the Relatives.
Another thing is that menus should be dynamic and based on where the user is within the taxonomy. Those menus should be changing and be specific to the page they are on. Not one all encompassing menu that leaves the users wondering where to click next, they don't need to see all that stuff.
Mega Menu Performance
Mega Menu Appearance
Have you ever been to a site and the menus are so complex that you find yourself spending way too much time trying to get around? No, there couldn't be websites out there like that, could there? You mean you saw a menu that had over 400+ links in it? And, it was being served on every single page of the site that way? Usability nightmare and an indexing challenge to say the least. If you don't have the PR to funnel throughout 400+ links you're running on empty. In fact, you're running on sand at that point. ;)
Typically I begin to worry when my surrounding navigational elements outweight my core relevancy elements. Mega Menus are at the top of my list when rebuilding taxonomy, those things have to go. In fact, that whole accompanying architecture needs to go. Rebuilt from the ground up to be more dynamic and in tune with the visitors "place" within the website. If that visitor came in through a specific search, I want the page they land on to be "very specific" and not cause them to hit that back button. Mega pages in general seem to have that effect on many. I know I back out if there is "too much" on the page.
The problem I see with removing a bunch of pages from the home page menu (drop down or not) is that those pages that are removed will have less PR. Yes, PR is not what it used to be but it still has some effect on ranking and particularly indexing.
I removed about 10 pages from my menu last year because I was getting close to the 100 page count and they dropped in PR by 1. I put them back in several months later and their PR came back. So I would only do this to pages that you don't care if they drop in PR.
|So I would only do this to pages that you don't care if they drop in PR. |
Did they also perform poorly with the drop in PR? I don't see that happening in most instances.
Last year when I was affected by the -950 penalty, the major change I made to my site was the left menu. I previously had the same menu on the side of every page. The top of my page was the main menu, and the side menu was links to the sub-categories. I changed my left menu so that each sub-category has its own left-hand menu. This, in effect, added more key words to my article pages. When my rankings came back I actually gained more traffic from these changes because of the added key words. My site never was affected by the penalty again...
If you have 5,000+ individual products in individual categories how are you supposed to divvy them up? Right now we use 26 links off of the main page to A-Z pages. Each letter has 20-500 links on it to each product. It's really the only thing that makes sense in our niche. This isn't Red Widget/Blue Widget/Green Widget, it's Red Widget/Large Thingamabob/Soft Gadget. There's really no way to combine them into anything smaller because they're so vastly different.
With all that said: we don't have a large laundry list of links on every page, just that one set of A-Z links from the main page. The subpages have links back to their letter, but that's it.
(Also: we've been in the -950 for about 2 years and have yet to find anything that works.)
Where it is believed that a long left hand menu is detrimental, is that menu code being placed at the end of the page coding and then positioned using CSS?
I was in the belief that SE's would look at the content that appeared first so the long menu, containing links that are off topic as far as the page is concerned, would not be a problem
|If you have 5,000+ individual products in individual categories how are you supposed to divvy them up? |
I'm not sure but there are always better ways to do things, always!
|Right now we use 26 links off of the main page to A-Z pages. |
A-Z pages? 20-500 links on each page?
|With all that said: we don't have a large laundry list of links on every page. |
Sounds like a laundry list. ;) Not on every page but on the second level of pages that may count. More than 100 links on a page would be a concern for me from a "usability" standpoint. More than 300+ links I think is the Kiss of Death for many sites. As a user, I surely don't want to have to wade through a page with a "Laundry List" of 300 or more links and you say you have upwards of 500?! That may not work and may be the reason why you are in your current situation.
|Also: we've been in the -950 for about 2 years and have yet to find anything that works. |
As mentioned above, a rebuild of the taxonomy might have been in order back when the symptoms started to appear. You could do it now too if things have not improved, what have you got to lose? And please, don't tell me that there is no other way, there is young man, there is! You must find it. Just because the competitors are doing it that way, doesn't mean that you have to. You can always set the new benchmark and then let them all flounder about while they try and figure out what you've done.
|Where it is believed that a long left hand menu is detrimental, is that menu code being placed at the end of the page coding and then positioned using CSS? |
While that addresses part of the challenge, it still leaves that laundry list of links within the "overall" content of the page. I feel that by using absolute positioning of those elements, you can focus the meaning of your content more efficiently by serving the primary content first and then other elements in their respective order. You want the page to degrade gracefully when CSS is off so the ordering of the elements is imperative from a usability standpoint. And of course an indexing one too. ;)
|I was in the belief that SE's would look at the content that appeared first so the long menu containing links that are off topic as far as the page is concerned, would not be a problem. |
Me too. And it usually isn't. But, when you get into the "Mega Menu" mindset, there may be some challenges there. That's why I feel menus should be dynamic and specific to the category you are in. Top level categories remain visible at all times. Sub level category links only come into play at those levels. If you have 5 categories, you'll have 5 main menus. If you have 50 sub-categories, you may have literally 50 sub-menus depending on the breadth of each category. You build it as they traverse the link path and only serve what is "specific" to that category. Did that make sense? I have to read these over and over to make sure I didn't confuse anyone, myself included. :)
"There's really no way to combine them into anything smaller because they're so vastly different."
5000 individual products are still easy. use the alphabet if you want but break it down. Sa-Se, Sf-Sr, Ss-Sz, stuff like that.
Google has said the 100 links a page thing forever. It's a good idea to think they aren't kidding. Having two or three pages on a site with 120 or 137 links isn't deadly, but having more than 100 links on a page, wherever they are, is to think Google has just been not serious for six years about pretty much the clearest suggestion they have ever made.
Looks like jake is looking at a site redesign..
Tedsters thread, which is a great reminder for a text to code ratio may not necessarily apply to my site since there is no mega menu, but it did get me looking at my layout, whcich always seemed a little heavy on navigation...
(Posted from a blackberry, gosh I luv technology)
|Which is a great reminder for a text to code ratio. |
Oooh, I was waiting for that phrase to come up. Many feel that the text to html ratio or text to code ratio is not a factor. I think otherwise. Also, once you come up with that percentage, then you need to determine what percentage of that is duplicate. Subtract that and you are left with the percentage of unique content, that is what counts. Minimizing the impact that a Mega Menu would have in this instance is of great concern for many.
Ever see the commercial with the State Troopers aiming their radar gun at a car dragging an anchor? Its funny! They start giggling when the radar registers 12 MPH and then an oil sludge promotion comes into play.
That's how I view websites with large navigation structures. Many times they become anchors in the true sense of the word. Most sites cannot handle the type of internal link architecture they are being bulked up with. It dilutes "everything". You cannot logically focus a website when the navigation structure remains "un-focused".
Now, the Wiki, what a piece of art that is. That is pure genius from my perspective. Google seems to agree too. Oh, and so do the other SEs, the few that are left. At some point, you become so large that navigation is no longer a factor, you rely on search and other on page methods to navigate through an array of taxonomy that is specific to your "experience". Its a rather "divine" concept if you ask me. :)
Ted, I think your right on. I had what could have been considered a "Mega Menu" and the site was 950'ed, after re-designing the entire link structure/menu to a more managable level most of my pages came back to their original positions.
I'm convinced this was a side effect of the Google Bomb fix that Google employed in Jan 2007.
oh oh oh, and plus, I was talking earlier about navigation menus showing up as the snippet in a site:example.com search.
How did that operator change your results after you started minimizing mega menu's?
I changed it side wide, got rid of the top nav menu, and now I can see what changes happen...
|5000 individual products are still easy. use the alphabet if you want but break it down. Sa-Se, Sf-Sr, Ss-Sz, stuff like that. |
But man would that look ugly fast :)
|Google has said the 100 links a page thing forever. It's a good idea to think they aren't kidding. Having two or three pages on a site with 120 or 137 links isn't deadly, but having more than 100 links on a page, wherever they are, is to think Google has just been not serious for six years about pretty much the clearest suggestion they have ever made. |
We've never had a problem getting crawled. All our pages get crawled. And we always ranked well for those pages... until -950 came along. But heck, if making the page all aesthetically ugly fixes the -950 problem then I guess it's worth it.
I have a site that is modeled on large news websites, lots of content, and maybe 20 items in the menus. PLUS, like most large news websites, we have say, 100-120 news items on the homepage with a bunch removed and a bunch added daily.
Would this be a problem for us, or is this specific to more thin-content sites?
wanderingmind, if you're not seeing a problem, it may not be a problem. You are talking about a relatively minimal menu, but a large assortment of ever changing links to fresh content. If googlebot comes by often enough and grabs all that fresh content, you're OK - it's not a "mega menu" that gets repeated throughout the site.
|...like most large news websites... |
I would just add, that emulating other sites, even industry leaders, is not always the best thing. I've often given in to ecommerce clients who say "but Amazon does it", only to watch Amazon STOP doing it within a few weeks or months.
You can be the business that comes up with the better idea and grab a major edge. Sometimes industry leaders have trouble shifting gears or even noticing that they have something going on that's less than optimal, shall we say.
mega menu, in a SEO point of view, would not be the same using a sitemap?
"But man would that look ugly fast :)"
Why? LOL, there is nothing ugly about Sa-Se compared to S-T.
If you use the alphabet, you aren't exactly aspiring to Shakespeare. Breaking the alphabet down into 50 units is easily understood and users would have no problem with it. It's not as ideal as "red widgets" and "green bodgets" but it's logical and Google would have no problems with it.
oh dear, I heard it so many times that now I feel sick when I look for something into Amazon
| This 71 message thread spans 3 pages: 71 (  2 3 ) > > |