Free Optimization Test Recommendation: The SEO Footer

March 25th, 2009

One of the great things about testing and optimization is that you can use it to challenge assumptions and keep your site design’s effectiveness up-to-date. No website element, no matter how pretty and well-designed, should be excused from an occasional test to make sure it’s giving you ROI.

In our fast-moving industry, what gave ROI six months ago might not today.

Take, for example, the design pattern I like to call the “SEO Footer.” This design element takes many forms, but essentially it’s using the footer of your website navigation to house keyword-rich anchor links as a means of increasing Search Engine Optimization. It’s a decent idea, and was extremely popular a few years ago, but I personally find them ugly, hard to scan, and hard to use.

Personal opinions aside, we’ve got a great candidate for an optimization test. Why? Because its impact on your site’s SEO is much less clear than it was a few years ago. And if it’s not giving you ROI on organic search traffic, guess what? It’s just cluttering up your site design and hurting your chances of converting customers.

I’m the first to admit I’m not an SEO expert, and the real SEO experts have written some good stuff on this topic, but in my research, I couldn’t see any evidence that sites are testing to see if this tactic is still “paying off” like it did 6, 12, or 18 months ago. So, there’s your free test recommendation.

“Wait! What Do I Test? And How Do I Interpret the Results?”

Run a simple test where some portion of your traffic sees a version of your footer stuffed with juicy keywords, and some portion doesn’t. Measure its impact on whatever conversion matters to you, or whatever conversion might rely most on intuitive navigation. Examples: Visiting a specific section of your site devoted to a business vertical and downloading a white paper, or browsing and finding a particular product and adding it to a shopping cart.

You may find the presence or absence of your SEO Footer has no impact on conversion. That’s OK, because you’re also going to dive into your analytics and/or your SEO reports and see if the absence of the footer hurt your rankings/traffic. If removing those extra keywords has positive or no impact, then you may as well remove it in the name of simple, clutter-free design. If removing it has negative impact, then you should optimize the existing design to make it look as good as possible, and resolve to test it again in six months.

Use this method to make every site element earn its right to be a part of your design, and you’ll optimize your way to ROI.

In A/B Testing, Improving Conversion, Search Engine Marketing

Add Your Comments

Comments (25)

J.F. Houpert
March 25th, 2009
11:17 am

I couldn’t agree more and with tools such as Google website optimizer it’s easier then ever to test site elements and gather precise data on what works and what doesn’t.
Linda Bustos
March 25th, 2009
1:10 pm

Personally I don’t mind footers with links – but I do believe footer links have long been devalued by search engines but as a user they don’t irk me much. I have seen retailers that will remain nameless but are HUGE with great search rankings using footers to link to their brands in low contrast to their backgrounds so these are DEFINITELY not there for navigation purposes

Personally it’s not something I would test, I’d rather find the best practice for SEO and do it. Suggestions are “page rank sculpting” – adding nofollow attributes to various links
http://www.getelastic.com/killer-seo-trick-only-1-of-online-retailers-use/
and fixing duplicate content problems like http://www.getelastic.com/too-many-urls-spoil-the-seo-fixing-a-common-ecommerce-duplicate-content-problem/
and creating SEO optimized category pages like http://www.getelastic.com/karmaloop-seo/

~Linda
Chris
March 25th, 2009
2:04 pm

It is easy to test for conversions, but how do you test for the SEO value of the footer links? Change it and wait 3 months to see if your ranking changed? Even if you had that kind of patience, how would you know it was *that* change that caused your rank change and not just a change in your competitors sites?
florida web design
March 25th, 2009
2:35 pm

I know for a fact that footer works, cause we had a site without any SEO opti at all, except for a footer like that, and it ranked well for all the footer links almost.
SEO News | Search Engine Optimzation Program Review
March 25th, 2009
3:43 pm

[...] Free Optimization Test Recommendation: The SEO Footer … [...]
Brendan Regan
March 25th, 2009
6:40 pm

@Chris: yes, you would “serial test” for SEO value. Since GWO and other tools don’t handle serial testing, you might have to do it as two separate tests, 1 to validate conversion impact, and 1 to validate SEO impact. On any serial test, you want to minimize other changes to your site to reduce the chances of “noise” in the data.
Scarves Shop
March 26th, 2009
5:00 am

Thanks for heads up. I think its worth testing for 1 page on your website that if you remove footer, how it affect SEO rankings.
Seattle Website Design
March 26th, 2009
11:50 am

Like this line, “No website element, no matter how pretty and well-designed, should be excused from an occasional test to make sure it’s giving you ROI”. All that glitters isn’t pretty.
roblogger
March 27th, 2009
8:31 am

nice tips. thanks an ciao from romania.
Thomas
March 30th, 2009
4:47 pm

I often use the footer to get in some text and to add any tags if my page is missing them. Also you need not wait moths to see results. Any time after your site is crawled you will immediately be able to see if your SERP’s have been affected.
florida web design
March 31st, 2009
10:06 am

I know for a fact that footer works, cause we had a site without any SEO opti at all, except for a footer like that, and it ranked well for all the footer links almost.
City Smart
March 31st, 2009
1:35 pm

hmm…split testing your SEO efforts can be extremely difficult & I’ve yet to find someone who can accurately do it consistently.

I might be missing something but with the example you’ve given above to split test this, the search engines wouldn’t have 2 indexed versions of your 1 page. If I remember correctly Google Website Optimizer is using Javascript, so I’m not sure you could test this?

Any real test couldn’t be run side by side, they would need to be 3 months vs. 3 months (which has additional factors that mess with the test).

As I mentioned I could be missing something, and would love to be corrected. If there is a split test idea for SEO – I’d love to see it!
Brendan Regan
March 31st, 2009
4:59 pm

@City Smart, you are correct that split-testing would be messy and/or impossible. I’m recommending serial testing, not split testing.
John
April 1st, 2009
12:42 am

Thanks, great job guys! What’s hot today? SEO is basically eye candy for search engines, too much time and effort for little return. A/B split testing it’s the surest results oriented way to turn even a mediocre site into money machine. If you’re marketing online, you can make much more money simply by improving your landing pages to incrementally increase your conversion rates. Even a small boost can mean a big return on your bottom line.
Larissa Reynolds
April 1st, 2009
1:52 pm

Hi Brendan,

WebDesignerWall had a post recently about the usability benefits of putting a sitemap (and sometimes company description) in the footer. That might be another variation to test, to see if you can maintain/improve SEO and still create a better user experience. Look forward to hearing the results from any tests you might run!

Here’s the link to the post:

http://www.webdesignerwall.com/trends/modern-sitemap-and-footer/
Larissa Reynolds
April 1st, 2009
2:33 pm

Here’s another link that I came across right after I commented above. Seems like everyone is interested in footers! More testing ideas:

http://www.uxbooth.com/blog/6-things-to-include-for-a-user-friendly-footer/
Brendan Regan
April 1st, 2009
8:10 pm

@all: i love that this has turned into a lively, multi-faceted discussion. One point of clarification is that I see a strong distinction between what I call an “SEO Footer” and what the UX community might call a “Fat Footer.” SEO Footers are just there for link juice, with little/no forethought given to the customer experience. Fat Footers (http://ui-patterns.com/pattern/FatFooter) are footers that have a lot of useful links in them. Take your pick, but I favor a “fat footer” or a standard footer approach, and if you can build in some link juice without damaging the customer experience, congrats!
Henry Philllips
April 7th, 2009
7:12 am

Since this article was published no less a site than the BBC (on its sports pages) has introduced a new link-heavy footer, although whether it would count as an SEO footer or a UX-approved “fat footer” is probably in the eye of the beholder (isn’t it all just down to a bit of styling anyhow?)

See http://news.bbc.co.uk/sport – look for the A-Z of sport at the bottom of the page.
Leadership
April 12th, 2009
10:24 am

I’ve been making minimal use of my footer, but think I’ll be giving this ’seo’ footer a try. I’m currently on a big link building campaign, so I won’t be able to test it properly – I’ll probably just ‘do it’.
Maltese SEO
April 14th, 2009
3:44 am

Creating a flat architecture, so that internal urls are reached easily by users and search engines is very important. However this should not be done in exchange of cluttering any web design. There are ways and means to implement this and by having loads of footer links, it’s not the ideal way.
Quinton Hoover
July 25th, 2009
12:47 pm

Having SEO footers is an old tactic.Search engines count them as spamming if over done..But this tachnique works like a magic for when you are targetting for low competitive or long tail keywords..
Molly Smith
August 25th, 2009
5:39 pm

It is a fine balance between user experience and SEO optimization. Having the footers for the sake of optimizing those keywords is fine, but there is a fine line when it becomes a truly bad experience with the clutter.
Oes Tsetnoc
September 15th, 2009
8:00 pm

I think that SEO footers should be designed with the user in mind. It is becoming blatantly obvious that some sites are geared just for the search engines when you have a nice graphic design up top, but then this gob of text near the bottom, text that would be much more useful for visitors if it were at the top instead.

It seems to be a controversy now though, as I have read lots of things recently saying that footers are now starting to lose their value in SEO.

~ Kristi
healthcare
October 4th, 2009
5:23 am

I’ll probably just ‘do it’.
Tanie Bilety lotnicze
November 18th, 2009
1:23 pm

Great topic.
Its really hard to test though what type of footer works. Id suggest makeing one that doesnt have too many links in it, most are permanent (ie dont change each time you get on the page) and dont look too spammy so as not to pass a manual review. Remember that a good footer helps in indexing which is very important.

Add Your Comments