Killer SEO Trick only 1% of Online Retailers Use
Stephan Spencer wrote an excellent article for Search Engine Land last week that explains how you can use the rel=nofollow attribute on your internal links to control the flow of “Page Rank” throughout your site. If this is all Greek-geek to you, I’ll explain in a moment what this means. But you should read this post because this is one white-hat SEO tactic that hasn’t been milked to death by all your online retail competitors. Out of Internet Retailer’s Hot 100 online stores, Stephan found only one using this technique, and even that store could go a bit further with it.
Page Rank (think Larry Page) is Google’s way of assigning authority to a web page. Your home page is likely to have the highest Page Rank because it’s linked to more often by other websites than your other pages. Page Rank flows between pages on your own site, flows in from other sites’ links to you and “leaks” through links to other sites. If you need more information on this, check out SEOMoz’ Whiteboard Friday on the subject.
Sculpting Page Rank is really plugging up leaks that don’t need to be there, and controlling the flow of “link juice” within your site, sending more juice to important pages like product pages and cutting off unnecessary pages (that you don’t need to rank in search engines) like contact, view cart, privacy policy, terms and conditions and so on.
When you selectively add rel=”nofollow” to links like “add to cart,” “buy now,” “submit,” for example, you tell search engines not to follow the link as they crawl your site and not to include the link in the overall (and highly complex) Page Rank calculation.
Stephan found that outdoor gear retailer Altrec was the only Hot 100 retailer that uses nofollow on sizes, expanded views, duplicate pages, and hot links which has won Altrec top placement on a number of keywords for category and product pages. Stephan found some product pages with Page Rank of 4, higher than many websites’ home pages! But Stephan suggested Altrec could go a bit further by nofollowing “Today’s Deal” and “Give Us Your Feedback” among others.
Sears, for example, has thousands of product pages. Some of its Web 2.0 goodies use javascript links which don’t pass Page Rank anyway, so not to worry. But it does follow links like “Add to Wishlist” — for every product page! NOT necessary. If you use the site: command in Google (to show you if a page is indexed), the URL for adding a Black Craftsman 4-Drawer Ball-Bearing Griplatch Utility Cart “Add to Wishlist” is indeed indexed. It doesn’t need to be.
site:http://www.sears.com/shc/s/p_10153_12605_00959756000P?vName=Tools&cName=Garage+%26+Tool+Storage&sName=Utility+%26+Mechanics+Carts#addToNewWishlist_
You can plug in site:http://www.yourecommercesite.com (use your domain instead here!) to discover what useless pages Google’s found of yours, too.
Here’s a nofollow checklist that is by no means exhaustive:
- Legal Notices
- Privacy Policy
- Order Tracking/Status
- FAQ / Customer Support / Help / Live Help
- Submit Review
- Testimonials
- Contact Us / Feedback
- About
- View Cart
- View All
- Add To Cart
- Add to Wishlist
- Tag This Item
- Click to Verify (VeriSign and HackerSafe seal)
- Register
- My Account / Sign In
- Enlarge Photo / Mulitple Views
- Size Chart
- Shipping Info
- Return Policy
- Careers
- Investor Relations
- Affiliates
- Click to Verify” VeriSign and HackerSafe seals
- Permalink (blog)
- Archive (blog)
It’s important to note that these pages can (and should) still show up for internal searches - customers seeking this information on your site - just not in the major search engines that recognize the nofollow attribute. We know that many online retailers’ internal search already excludes this and users sometimes resort to external search engines to find the information they need, (like “PuppyPalace + international shipping). So if you block off important pages through Page Rank sculpting, be sure they can still be found with internal search.
When you nofollow these types of links, you’ll apply the attribute to the site wide links, but that doesn’t mean you can’t still pass real links to certain pages and get them indexed, either. You can pass real link juice through a site map, or if you have a separate corporate site, you can for example link to your affiliate program page on your ecommerce site from the About Us page on your corporate site.
Though this all sounds complicated, it’s one of the easiest and ways to optimize your site for search engines and can really help you capitalize on long-tail traffic and gives you a competitive advantage.
Hi Linda, the link to Stephan’s post from your blog takes me to your post. Just wanted to point out.
Thanks for the tip!
Great post Linda and very timely…
These are changes I plan on making to our ecomm. site in the next couple weeks.
One statement that you made confused me a bit:
“If you use the site: command in Google (to show you if a page is indexed), the URL for adding a Black Craftsman 4-Drawer Ball-Bearing Griplatch Utility Cart “Add to Wishlist” is indeed indexed. It doesn’t need to be.”
A page being “indexed” and a page passing “link juice” are two different concepts.
The NoIndex and NoFollow metatags are inherently different where one will keep a page out of Google’s index, while the other will just keep Google from following the link for the purpose of page strength purposes.
Not sure if this is what you meant, or if I just read it wrong! (i do that often)
Great work on this blog…I find myself visiting often.
Thank you,
mb
@ emon,
Thanks for pointing that out. I can’t get away with trying to hoard all our own link juice can I? Just kidding.
@ Mike,
You are right about the noindex/nofollow difference, and a noindex page can still accumulate and pass Page Rank it just won’t be delivered in results, whereas an indexed page can have nofollowed links that don’t pass Page Rank.
I was under the impression that nofollowing all links pointing to the page would mean that Google also won’t use those links for discovery. Of course, if you apply the attribute after-the-fact, you might have these pages stay in the index and that’s not a bad thing.
From Matt Cutts’ quote in the SearchEngineLand article:
“There’s no stigma to using nofollow, even on your own internal links; for Google, nofollow’ed links are dropped out of our link graph; we don’t even use such links for discovery. By the way, the nofollow meta tag does that same thing, but at a page level.”
So if you’re starting a site from scratch and you nofollow all links pointing to wishlist pages (and others) they won’t get indexed unless some other page links directly to them. (This is how I understood the concept).
Of course, existing sites will have a ton of pages and the site:operator can give you ideas for what stuff you don’t need to link to.
Please let me know if I misunderstood how this words :D
I believe what we both said is correct. :)
Keep the great posts coming!
We started doing this a few months ago on one of my sites, and I’ve definately seen a boost in ranking for deep links such as product pages.
I’ve also found that showing as many products as possible on the product category pages has helped get them crawled quickly. Requiring Google to crawl through page 1, 2, 3 and so on to find the product pages kills the PageRank flow.
@ Justin,
Great tip on the show all, I think this is good for usability too. I don’t know WHY anyone would want to click through each page when they can see all at a glance. I always select “View All” anyway and get frustrated with sites that don’t use it :D
This is the sort of tip that will give you the slight edge you need to outrank the competitor’s above you in the SERPs.
I’ve used this technique on a few of my real estate client’s sites and actually saw a slight increase in rankings almost immediately for nearly 50% of my short and long tail keywords.
This is an interesting tip, and one that I admit not being aware of even though I am familiar with the nofollow tag.
I do have a disconnect in the explanation though. How does excluding pages like “Contact Us” help your pagerank for other pages? I’ve re-read the article but I’m missing the relationship. Perhaps I simply need to accept the black box that is Google’s pagerank algorithm.
@ Will,
Thanks for joining the conversation :D
When you link to your top level navigation items from every page on your site, you send a lot of link strength to them that they don’t need (you don’t expect these pages to pull in search traffic). Since Page Rank is distributed to all links on a given page, by cutting off less important pages, you send more link juice to the pages you want.
For example, if your home page has 50 links, each link will receive a 2% share of Page Rank. If you block off 25 of these links through nofollow attributes, each link gets 4% or twice as much. Now if that happens across all your pages, your category and product pages start receiving much more link juice in the aggregate.
Yep this is confusing stuff :D
This is interesting and something to keep an eye on, but at the same time, there seems to be a lot of assuming that we know exactly how Page Rank works.
The last time I checked the algorithm isn’t public, so we may assume this helps, but is there any proof?
I mean it seems that if this was 100% true, then the fewer pages you have, the better their rankings. So before everyone was told to generate as much legit content as possible, and now linking to that content is hurting the pages PR?
Hi Tom, thanks for joining in.
The mysterious Page Rank formula is not publicly known, but I do believe there are patents that are public, and SEO expert conversations with Matt Cutts (of the Google spam team) and Matt Cutts endorses this technique as a way to ethically optimize one’s site.
Page Rank is also assigned according to how authoritative your content is, creating more pages of good content that others link to increases your overall inbound link count for your domain - which can increase home page PR which gets recycled somehow throughout the site. Yes it’s all quite confusing. But we do know that page rank flows between pages and can leak, when you plug up the useless leaks to preserve juice for the pages you want to strengthen.
Great info Linda,
I have one question though can you apply nofollow to bookmark buttons for example the addthis button?
thanks
@ Simon
Yup, you can nofollow an addthis button too.
Hi Jason,
I can’t seem to get this to work on the following code , any ideas? thanks.
addthis_url = location.href;
addthis_title = document.title;
addthis_pub = ‘mydomainname’;