Forum Moderators: Robert Charlton & goodroi
Then we thought we'd get smart and lift the conversion rates, by building in some unique functionality onto our main sites. Search filters, sorting columns with added internal links, images and so on. Our links inflated on these pages, sometimes to around 200 because of this, and the links to text ratio increased in favour of the links.
[ During this time we messed up our site architecture, creating duplicate content with conflicting indexing paths and all of the sites tanked off of G's search index. That situation is now fixed. ]
Just recently part of our network showed great signs of life, and bingo, guess what, all the sites that were plain simple shot to the top of the results. But all the sites with added functionality, which would have been very useful to users, and provide excellent conversion rates to us on lower visitations, are nowhere to be seen on G.
Does added functionality conflict with plain simple SEO, or can the two co exist?
What's been your experiences on this and what limits can you push.
I also agree with your suspicion that too many links on a page is a problem. The algo skews too far toward the Tower of Babel created by all that anchor text, and your relevance signals for the page get watered down. I started a thread about this a few years ago with regard to DHTML Dropdown Menus [webmasterworld.com], but it applies to any situation with a plethora of links.
Also these Library threads below address what you can do to create a solid Information Architecture, instead of just flooding the page template with hundreds of navigation links:
Information Architecture for the Small Site - part 1 [webmasterworld.com]
Information Architecture for the Small Site - part 2 [webmasterworld.com]
Putting information architecture into practice [webmasterworld.com]
Bottom line for visitors --
A web page is not usually the same thing as a software application. If someone is using a bit of software regularly, then they will take the time to learn at least some of the nooks and crannies. But if your website is something used once in a while, you really need to simplify the interface.
Bottom line for Google --
The relevance "signal" for any URL is essential. Simplicity boosts that signal and dials down the noise.
I'm not convinced that Google's quote of "build sites for users, not Google" is entirely true. The context needs to be well understood before pursuing this strategy without thought.
I became suspicious when I observed some top branded sites with great functionality, opting for dumb and simple pages that ranked well in highly competitive sectors.
Whilst restoring a network of our sites from duplicate content issues, we have persisted and invested heavily in a branded network of sites that has great on page functionality - but the results on Google for these sites were very disappointing - nowhere near their former glory days when the sites produced great results.
Great functionality, meant we built the site for users, not Google. This was validated by the users, since the conversion rate to traffic ratios were very high - way higher than anyone in our industry sector. But without lot's of new traffic, it's pointless working solely on conversions.
These established sites of over 3 years had PR5 and 6's, with good on topic IBL's from .edu , .gov and quality sites, but even on obscure keywords, would not rank consistantly.
In stunning disbelief, we observed three [ out of four .. the latter is not yet performing? ] of our different foreign language sites, with very basic content layouts, return to the serps and rank well with good traffic. There was also minimal inbound links.
So, to see if traffic can be restored, we have returned our flagship site to a very basic layout, the same as the foreign language sites, to see if it will work. Time will tell, if we haven't been "footprinted".
The keys to the simple site's success seem to be:
1. Site wide architecture [ no dupe content and heirarchal link theming throughout]
2. Original page content on topic, simple text. [ removal of all product search filters ]
3. Quality inbound links [ minimal in our case ]
Has anyone had better success by focusing on building for Google versus building for users?
[edited by: tedster at 3:43 am (utc) on Dec. 29, 2006]
There is a solution: in my view people should stop bashing Flash for its low ranking ability - it can be used on XHMTL pages with a result that is frankly NEVER going to change your ranking if you add or change its sturcture as actionscript is in many many ways the same as javascript - freedom for those that can use it well.
Example 1: too many navigation links
This puts too much anchor text on all the pages and it seems to dull Google's ability to make strong relevance determinations. But it does the pretty much the same thing for end users -- it overwhelms them with choices and they often choose to leave.
I've seen it over and over -- too much nav is a problem, and that includes those icky DHTML dropdown menus that were adopted so frequently in recent years. I my view they usually are a lousy and lazy replacement for a well executed information architecture.
Too many nav links is like that guy on the bus that gets your attention in some simple way, but then just won't shut up. Several Monty Python skits come to mind.
Example #2: the hodge-podge page
Sometimes a page comes out of the development process with no clear topic -- except maybe for "FAQ", which is usually tecchie talk for "junk drawer". Dead giveaway that I've got one of these hodge-podge pages is that I can't come up with a single H1 element that contains a respectably targeted keyword.
Odds are that Google will not find a decent search term to rank this page for -- but odds are also that users will seldom find a reason to click on a link to the page, or to stay there either. It's much better to create bite-sized pages around conceptually focused information.
Is that building for Google? Well, sort of. The fact that G doesn't know what to do with a hodge-podge page is certainly a motivator for change. But again, I see in the stats that visitors appreciate the clarity as well.
Simplicity is the key to clarity. But it takes a lot more work to arrive at functional simplicity than just to fling content into a template.
[edited by: tedster at 4:17 am (utc) on Dec. 29, 2006]
My full process is a lot like using Google as a usability analyzer
I like this powerful, yet subtle way, of inverting the usability issue with Google to meet user requirements. It kinda replicates the two perspectives into one uniform approach. A very useful sentence Tedster. Thanks.
How important is the use of CSS in the background, when establishing the usability foundations?
[edited by: Whitey at 5:30 am (utc) on Dec. 29, 2006]
I'm kinda perplexed as to why 3 out of 4 of our foreign language sites have kicked and one hasn't, even though they are identical, except for the language content.
