Search and post jobs for the Multichannel Merchant. Including jobs for brand & agency marketers, e-commerce, catalog marketers, ops & fulfillment, direct marketing and more.
The
category and subcollection pages are not making it into the search
engines at all — not because of their spider unfriendly URLs, but
because they are being specifically blocked through “disallow”
directives in the site's robots.txt file. Robots.txt is the place where
you can give commands to Googlebot and the other spiders, such as “stay
away from this directory” or “stay away from this file type.”
Looking
at the site's robots.txt file, I see that category pages,
sub-collection pages and product information pages are all being
disallowed. You might wonder why this is the case. Wouldn't Red
Envelope want these pages indexed and ranked, despite any inherent
search engine unfriendliness? Well, in this case, the answer is no.
That's
because Red Envelope has alternate pages for the spiders to index,
using a technology called SearchDex. SearchDex autogenerates thousands
(more precisely, 2,250, according to Google search results) of sitemap
pages. These doorway pages (such as the one at http://www.redenvelope.com/giftcatalog/Ccat10095.jsp) are built specifically to lead spiders to product-level content pages.
These
SearchDex “Ccat” pages may appear to be full of meaty, keyword-rich
content, but upon closer examination of the content, it is quite
apparent that the content was not written by a human. For example,
consider this fine prose in the second sentence of the first paragraph
on the aforementioned Ccat10095.jsp: “Our men's accessories range from
men's fashion accessories to men's leather accessories, which are
reasonably priced and unique to RedEnvelope.com.” Yuck!
Or consider the lead sentences on http://www.redenvelope.com/giftcatalog/Ccat10629.jsp:
“Great Christmas presents make holidays magic. Unique Christmas
presents from our collection of our newest gifts this Christmas season
will bring good cheer.”
It's
throwaway copy from the reader's perspective, but certainly dense with
keywords: 10 occurrences of “accessories” and “men's” in the first
paragraph of the former example, seven occurrences of either “Christmas
presents” or “Christmas present” in the second example.
There
are names for this black hat SEO tactic, none of which are
complimentary: “keyword stuffing,” “spamglish,” and “doorway page” are
just three that come to mind. This is a search engine ban waiting to
happen.
Also
in the aforementioned paragraph on Ccat10629.jsp, the words “my
Christmas presents” are actually wrapped within heading tags, yet that
fact is hidden from the user. The headings are given the exact same
font, style, and treatment as the rest of the paragraph copy, so they
are indistinguishable from the surrounding text and buried within the
paragraph.
Links
within the paragraph copy are hidden in the same way. Clearly, this was
done only for search engines and not for humans. This is definitely the
sort of thing that Google's automated algorithms seek to detect and
penalize.
The
title tags are similarly keyword-stuffed. A good rule of thumb with
title tags is not to repeat a word three times and not to repeat more
than two words. In the title tag of the aforementioned page
(Ccat10629.jsp), “Christmas presents” is repeated twice, “gifts” is
repeated three times, and then “gift.” Furthermore, the title spans 17
words — too long. I would go for a dozen words or less.
Looking
again at the cached version of the home page (the one that Googlebot
was given), I see that the majority of links on that page are wasted,
because they link to category and subcollection pages that are being
disallowed. Where are the links to the SearchDex pages?
There's
only one SearchDex link — to the top page of the SearchDex sitemap. And
that's quietly tucked away in the copyright line at the bottom of the
page, since the linked page is not really meant for human consumption,
only for spiders. There are no graphics on this sitemap page; it is a
page chock-full of text links to various SearchDex Ccat pages.
Links
contained on the home page along with their anchor text count heavily
toward SEO. For instance, the “jewelry” text link would, in normal
circumstances, help the linked page rank well for jewelry-related
searches. That's because the search engines associate the anchor text
with the page being linked to. Not so here; these navigation links are
of no value because of the disallow.
If
the “jewelry” category page weren't disallowed, it would be unlikely to
rank well due to the lack of text content on the page. Contained on
this page, however, are text links to search results pages. Search
results pages can make for good search engine fodder, but fewer than
100 of these search results pages are making it into Google's index.
And nearly all of those are in the supplemental index — an indicator
that they are unlikely to rank well in all but obscure queries.
The
logo in the top left on all the pages across the site (with the
exception of the SearchDex pages) links to the home page — but using
the spider unfriendly URL complete with session ID, rather than http://www.redenvelope.com/.
Thus these links pass PageRank to a different version of the home page
rather than reinforcing the PageRank of the true home page.
Although
you wouldn't be able to tell this from the Google toolbar (due to the
unique session ID-containing URL you will have been redirected to upon
visiting), RedEnvelope.com has a respectable home page PageRank. By
using the “PageRank Lookup” tool from SEOChat.com, I was able to determine that the home page scores a 6 out of 10.
But
since PageRank is on a logarithmic scale, 6 is not as good as you may
think — a 7 or an 8 would be much better. SEOChat's “PageRank Search”
tool reveals something rather alarming: The majority of the SearchDex
auto-generated pages score 0 out of 10, particularly at the product
level. A number of Ccat pages have a PageRank 2 or 3, and only a few
have a PageRank 4.
Yahoo Site Explorer (siteexplorer.search.yahoo.com)
reveals quite a healthy set of inlinks — from blogs, shopping sites,
news articles, directories, and so on. Yahoo counts nearly 20,000
inlinks (excluding internal links). With some re-architecting of the
site, this “link juice” could really be much more effectively leveraged
across Red Envelope's site. That — along with rewriting the URLs to
eliminate session IDs and “stop characters” (ampersands, equal signs,
question marks) from the URLs; discontinuing the questionable SEO
tactics of doorway pages and hidden links; and adding meaty content —
should have a profound impact on Red Envelope's rankings and search
traffic.
Would you like to have your Website critiqued by our panel of pros?
If so, e-mail Mark Del Franco (mark.delfranco@penton.com) with the name and URL of your Website and some basic information about your target audience.
Please include “Website Critique” in the subject line of your e-mail.
