In a Nutshell: This is a comparison of two product pages. The first one is from JC Penney’s website and the URL is too big to fit into this nutshell, and the second is convenient, the product page of the SEO Best Practices book I wrote. In the text below I’ll give draw a parallel between these two pages and point out the elements needed for any product page.
It’s quite a long read, so get comfortable and grab something to drink. I intended on placing this as an ebook so as to lure you to sign up… but decided to leave it out in the open like this, so if you like the material you can sign up for updates either on this site or head over to eBooksAndTutorials.com and sign up there.
1. Introduction
A few days ago the New York Times published a story on JC Penny’s recent SEO blunder. The crux of the story is that nobody is too big to make a big mistake, and nobody is too big to get punished by Google. The underlining principle is that unethical business conduct does not yield lasting results, and as such is not a wise strategy for building lasting success.
So what really happened with JC Penny? Well, in a nutshell, it seems that they hired an SEO company that has been building links using some black hat SEO tactics, which resulted in over 2000 links leading back to JC Penny’s homepage on keywords such as bedding, dresses, area rugs and so on.
This link building hack went so far that JCP was on the top of SERPs even for Samsonite carry on luggage, ahead of Samsonite.com itself. They ruled the SERPs for several months, and the holiday season “coincided” with this rule, so JCP made tons of money.
How much?
Google keyword estimator registers 11 million “dresses” queries. Knowing that the first position brings 34% clicks, this translates in roughly 4 million monthly visits to JCP. If their site converts the average 1% into buying customers, this means that in just one month, thanks to this link building hack, JCP got about 40.000 buyers.
The average price of JCP’s dresses is about $50, give or take a few bucks. So this traffic surge helped JCP to make $2M in monthly sales from dresses alone. Considering that they also ruled the cases, furniture, accessories and what not, the monthly sales mount up pretty quickly and we can easily understand what it means to be #1 on SERPs in raw dollar figures.
But, as the word goes, no evil deed goes unpunished, so, conveniently after the holiday shopping madness ended, Google finally caught up with JCP, and in literally 2 hours their positioning was re-evaluated, so instead of the comfortable #1, JCP dropped (out of SERPs) down to #60 and even #80 for some keywords.
Of course, JCP immediately fired their SEO company (that got a lot of hype thanks to this linking hack), promised to take down the purchased 2000 links per keyword, and of course, Google came out to save the day once again with their new algorithm update at the beginning of February 2011.
So I dare ask, did this really solve the problems for JCP?
In the text below I will compare a sample page from JCP’s heavily funded website, and compare it to a small startup site developed with virtually no expenses except for time, good will and a solid list of design and SEO best practices for e-commerce sites. The analysis is based on information gathered from www.seo-browser.com as a free and standard tool as a part of the on-page search engine optimization.
2. JCP: How NOT to do eCommerce SEO
2.1 Where Did JCP Go Wrong
JCP made a conscious (wrong) decision long before they hired their SEO company. They decided to build their eCommerce e site without following some common sense best practices, and to make things worse, whoever runs JCPs online actions missed the nature of search marketing altogether. What do I mean by this?
First of all, JCP’s site looks like a nicely painted picture over a terrible and flimsy coding structure. Second, search marketing is all about good content, and their poorly coded site offers none of it.
This poorly made site probably forced JCP to go for black hat SEO because white hat means going back to the website and actually making it search engine friendly. With these two major points out in the open, let’s go through the pointers as to why JCP’s site ruins its own sales.
I picked a dress product page just to stay on the subject matter of dresses. Here’s the link to the page, yes, I know, it’s quite messy and long (not an active link):
https://www2.jcpenney.com/jcp/X6.aspx?GrpTyp=PRD&ItemID=19449db&submit%20search.y=0&Ntt=new+year+dress&SearchString=new+year+dress&Ne=840+4+877+878+5+961+6+29+3+598+11+15+12+506+10+23+585+596+1031+8+18+904+903+969+949+833&hdnOnGo=true&NOffset=2&submit%20search.x=0&Nao=0&N=4294959029&SO=0&PSO=0&cmAMS_T=XGN5&cmAMS_C=MERCH&cmAMS_Z=XGN5TOPOFRESULTS&CmCatId=searchresults
2.2 JCP Coding Issues
JCP’s site is ASP-based, and heavily relies on JavaScript for functionality and content development. From an SEO perspective, whenever we see a JavaScript (or Flash, for that matter) powered websites, we get the chills. Why? Because searchbots cannot handle/index JS and Flash. So this is the very first issue with JCP’s site: it is hard to index its content.
Further on, the site’s metadata is nonexistent. There is no metadescription, which is the snippet of text that by default shows up in SERPs. The lack of metadescription combined with a JavaScript content handler means Google will have no text to put on SERPs, which greatly reduces the click-through rate. The theoretical 34% CTR goes for well optimized sites, and from a snippet-perspective, JCP’s site is both unoptimized and quite unfriendly to search engines.
Next, the webpages have no canonical URL, which makes it difficult for word-of-mouth and social advertizing, and also makes it harder for indexing purposes because the bots will have no way to know if this page was indexed. Instead, URLs are system-generated, long, non-descriptive… and plain ugly.
According to seo-browser.com, the bot-visible text on a sample JCP page counts only 101 words, which is far beneath the threshold for indexing, which is 250 words. The reason for this minimum words count is that pages with very little text get ignored by Google’s indexer because the rule of thumb is that you can’t really make a valuable contribution in less than 100 words.
The only exclusion to this rule is Google’s indexing of our Twitter activity. HOWEVER, looking at the HTML source code, there is an alarming issue, namely, there’s nothing between the and tags. That’s right, the Body section of the sample page is empty. Here is a short excerpt of the surrounding code of the Body tags:
/TR><TR><TD HEIGHT=15 ALIGN=RIGHT><BODY></BODY></TD></TR><TR><TD COLSPAN=
I’d think tables gave way when CSS got popular, but it seems I’m wrong. JCPs site is full of tables. Also, we’d all guess that multibillion dollar companies would have no problems to cash out for a site upgrade, but it seems that’s not the case with JCP. But I digress.
The page I reviewed had 15 images of which 11 had ALT tags, and at first glance, we’d say OK, at least they have the ALT… but once we get to the View Source, we quickly realize that the ALT tags are useless: Add to cart, Share on Facebook, Email a Friend, Pick a color and so on.
ALT tags should be descriptive so that searchbots can understand what the image is about. Also, people with impaired vision who rely on text-to-speech browsers rely on properly formulated ALT tags to better understand what the page is about.
Lastly, the webpage did not have a language metatag, which is automatically generated by most web design platforms. How JCP managed to launch a website without the basic metatags is beyond me, and beyond the focus of this post.
2.3 JCP Searchbot-Level View
Websites are more than what meets the eye. This is especially true if you’re seeing things through Google’s eyes. For search marketing, we better make all the efforts needed so that Google gets a good idea of what the webpage is all about. Fail this and your website will fall off SERPs… unless you use JCP’s black hat tactics, which as we’ve seen from the NY Times article, won’t do you good for long.
So, how do searchbots see the JCP test page? Ideally, the web page should offer the Googlebot information about the dress: model, material, color, style, available sizes, matching accessories and so on. Also, it should have a text explaining that the dress is suitable for both casual and more formal outings, usually best for indoor events (we’re still talking Winter season sales) and so on.
At last, Googlebot sees only JavaScript code, and gets stuck in the Me no comprende JavaScript loop, and here’s what we’d all see at JCP if we saw the world through Google’s eyes (emphasis mine):
[go back to jcp.com home page] find a store store ads store services track order help 1-800-322-1189 gifts + registry coupons & offers [Sign up for mobile coupons & offers] [Sign up for store & online offers] join us [Join us on Facebook] [Follow us on Twitter] women
home store
men
jcpteen
kids
baby
shoes
handbags + accessories
jewelry + watches
patio
clearance
[American Living] [Sephora] [Liz Claiborne] [Sports Fan Shop] [go to shopping bag] shopping bag 0 items $0.00
No-Script:
Javascript is disabled in your browser. Javascript is used to provide you with the best shopping experience.
Please enable Javascript on your browser and refresh this page before continuing. How to enable Javascript…
[view gift list] [continue adding gifts] [home page] > kids Bisou Bisou® Knot Front Dress-Black-Plus Sizes [spacer.gif] [print this page] print this page [Product Image] Black [Black] $39.99
Original $70.00
[bisou_bisou_logo.jpg]
Work this sassy little black dress from day to night.
V-neckline knot detail at shoulder empire waist polyester/spandex washable imported
Refer to our Bisou Bisou size chart to help determine your size.
[size_chart_btn_pp_062410.jpg] [ ] [out of stock] Pick a size 16W 18W 20W 22W 24W
No-Script:
Pick a color
No-Script:
quantity view your shopping bag [Share on Facebook] [Add to MySpace] [Add to Stylehive] [Add to Kaboodle] [email to a friend] [spacer.gif] [view gift list] [continue adding gifts] Customer Reviews https://www2.jcpenney.com/jcp/X6.aspx?GrpTyp=PRD&ItemID=19449db&submit search.y=0&Ntt=new+year+dress&SearchString=new+year+dress&Ne=840+4+877+878+5+961+6+29+3+598+11+15+12+506+10+23+585+596+1031+8+18+904+903+969+949+833&hdnOnGo=true&NOffset=2&submit search.x=0&Nao=0&N=4294959029&SO=0&PSO=0&cmAMS_T=XGN5&cmAMS_C=MERCH&cmAMS_Z=XGN5TOPOFRESULTS&CmCatId=searchresults DeptID=42249&CatID=42249
No-Script:
[close] email a friend Type the email addresses for up to 3 friends. They will receive a message with a picture of this item and a link to jcp.com to view. your email address: your name: friends’ email address personal message: (optional)
No-Script:
home | other products & services | store locator | gift cards | para ayuda | credit card | about us | contact us | copyright privacy [product view] black
As you can see, there’s not much to see for Googlebot at JCP’s test page down at the plain HTML level. There is no text for Google to index, and logic would suggest that such a page simply cannot rank well with text-hungry search engines.
2.4 JCP Site Analysis Summary
The JCP’s sample product page looks nice for users, but is not optimized for search engines. The coding is terrible, the URL structure is auto-generated and not memorable, there is no descriptive text and there is no useful metadata on the page. All in all, this kind of website can only rank well if they opt out for black hat SEO, which is exactly what they did… the problem is that no bad deed goes unpunished, and in JCP’s case, they’re down at #80 in SERPs, which is practically as good as being sandboxed. They probably spent huge amounts of cash in link purchases and now all that money is out the window.
3. How eCommerce SEO Ought To Be Done
Now let’s quickly take a look at www.ebooksandtutorials.com, a small startup website as a good example of a product page that is both user friendly and search engine friendly. I should note that this is a site that I created using WordPress as a platform with a handful of free plugins. The only expense in this project is the domain registration, hosting plan and my free time. I’ll tell you later why I picked my site and not somebody else’s site.
Now, let’s move on to the analysis.
3.1 eBooks And Tutorials Website Coding
WordPress is very Search Engine Friendly out of the box, but there are some steps to make it 99% SEF. I can’t say it’s 99% friendly because Google always changes the algorithm, and the bottom line is that only Google’s engineers can make a 100% search engine friendly platform. But then that would be cheating.
So, as WP is very search engine friendly, with a few plugins I easily got to set up the Google/Yahoo/Bing verification codes which is very important for traffic analysis. The metadescription entry is present, and hand-written too, so I have full control over what SERPs display next to the link to my product page. The metakeywords tag is automatically generated based on the words often mentioned in the text.
This metatag is not important for SERP positioning, but unlike JCP’s platform, WP easily outperforms and over-delivers.
Further on, my product page has a canonical URL, which makes it easy for searchbots to keep track of what’s indexed and what is not… and also having canonical URL that is keyword-rich and memorable is very good for word-of-mouth advertizing over the social networks.
The Text to Code ratio is 24.4%, in comparison to JCP’s 0.7% text in the entire HTML source code. This ratio is important in two ways: code optimization, and text importance. Usually, well-designed pages are light on code and have enough text to get to 20-30% ratio for top performing sites. This however depends on page content, so if there are tables the ratio can drop a bit, but the rule of thumb is to strive for the 20-30% bracket.
Next, there are 19 images, all with ALT tags, and granted, there are images like Next, Add to Cart and so on, but the main point is that the product-related images have descriptive ALT tags so Google knows what the image is about.
The Body Word Count is 1358, which is a bit over the top at first glance since the word on the net is that Google truncates/stops indexing of texts at the 1000th word. The logic behind this is that whatever was worth of mentioning was mentioned in the first 1000 words, and also Google wouldn’t like to store all that data just because folks like me love writing long posts.
However, knowing the WP coding structure, it’s obvious that there are a lot of extra words from all the widgets, sidebar, internal linking, user comments and so on. Actually, the text itself is 700 words, which is spot-on in the goldie-lock zone for web texts (unlike the current text you’re reading, that counts over 3000 words).
3.2 Searchbot-Level View
Now, let’s take a look at how Googlebot sees the sales page on my website (read it if you will, but the point is to make the difference between the level of clarity about the page’s content):
Home
Products
About Us
Download Page
eBooks and Tutorials Everything you need to know Computers and IT
Health
Holidays
Marketing
Personal Growth
Products
SEO Best Practices a No Nonsence Step by Step Guide for Focused Traffic
Categorized Under: SEO
SEO Best Practices a no nonsence, step by step guide for focused traffic is a practical book that will take you through the best practices of SEO so that you can quickly learn what needs to be done, and apply it to your website as soon as possible. The content is equivalent to a 10-hours SEO consultancy session with a top SEO professional. The only difference is that such a consultation costs over $2000. This book is way less expensive and just as valuable…
As you can see, the Body tag is very close to the actual start of the site that Googlebot sees. Of course, there are plenty of additions before, but Google skips all code; entries and focuses only on the meta tags (omitting the keyword meta of course) and the body text.
3.3 eBooks And Tutorials Site Analysis Summary
As we saw sofar, a smart approach to developing my website has helped me tick most of the checkboxes of the imaginary SEO checklist: the test page has all the needed metadata, the URL is canonical and keyword-rich, the web text is just the right length and fully indexable, the text to code ratio is in the prescribed boundaries, the internal linking is well made and all that is well within the tolerable HTML file size (39KB) and within the normal loading time (1.2 seconds).
As I promised above, here’s why I picked my site to compare it against JCP’s site: I know exactly how much time it took a single man to build this site, that there were no expenses other than my time, so basically, with the right know-how, anyone can make a search engine friendly ecommerce site.
4. Conclusion
J. C. Penney built an ecommerce site that is not optimized for the user and search engine friendliness and as such it cannot rank well in search engine result pages, despite all the money they probably threw in developing the site and funding their SEO efforts. On the otherhand, my shoestring-budget site vastly outperforms JCP’s site in search engine friendliness and user friendliness.
There is a great principle that we can all take home here: SEO/SEM is not about having the bucks, but having the brains. White hat SEO offers us the chance to beat industry giants by following the SEO best practices and be patient with the results, because once our site gets picked up by the masses the benefits will last much, much longer than the short burn-and-bust results from black hat SEO.
Need help with your site?
