A first pass look certainly seems to indicate that may be the case!
The now famous New York Times story on JCPenney’s questionable linking tactics resulted in what can only be described as extraordinary measures by Google to “correct” the retailer’s high rankings. Overnight the company lost its strong position across tens of thousands of keywords and, presumably, a massive proportion of its traffic from search engines.
Fast-forward three months later. seoClarity’s exclusive daily analysis shows that the retailer is out of Google’s ‘penalty box’ and again moving up the rankings and regaining lost rankings. So how exactly did JCPenney turn their situation around and what can be learned from it?
We set up several alerts to observe JCPenney’s performance on keywords where it held a strong position before the NYT story hit newsstands. On May 12th, almost precisely 3 months to the day when Google assessed its penalty, our alerts went off by the hundreds. We had observed some changes here and there over the past few months, but over 900 alerts on a single day meant something significant had happened.
Whatever has changed, it is the result of actions completely unique to JCPenney. In the normal day-to-day of changing links, content updates and Google algorithm tests, fluctuations in the top 10 positions are common – even more so in positions 11 – 100. This particular change was unique because the keyword competition shifted to accommodate JCPenney moving up without any other significant movement in competitors’ rankings. The competitors had simply ‘moved aside,’ where normally we would expect to see a major shakeup typical of a global algorithm change.
We selected a sample of over 2,000 keywords on which JCP ranked in the Top 100 for each of the key dates in this saga – February 10, (before the penalty got applied), May 10, and May 11 (the day after the penalty was apparently lifted). The breakdown of the keywords by word count is presented below. Our sample was distributed to allow us to observe significant changes for both head terms (1- and 2- word search queries) and long tail phrases (3+ words):
|Word Count||Count of Keywords|
This graph shows the count of keywords that JCPenney had in the Top Position, Top 3 and Top 10 across the three dates for the selected keywords (plotted on a logarithmic scale to make it easier to see the difference).
Notice the big jump from 24 to 899 keywords in the Top 10? The same jump appears regardless of whether the keyword is a head term or long-tail. Competitive sites did not show the same pattern in either a significantly positive or negative direction.
So far, we had confirmation of the following:
- A substantial movement had happened in JCP’s rankings
- The movement was limited to JCP and not an overall update to the algorithm.
What changes did JCPenney make to its site?
No doubt JCPenney has been hard at work to resolve the issue since the penalty was applied. The fact that their rankings are moving up again is a strong indication that whatever they did has helped. Rather than obsess over how much their rankings had changed, we were more interested in learning what JCPenney had done to get the penalty lifted. We had two places to look – their link graph (which is what originally got the site into trouble) and on-site changes. Cleaning out links was the obvious first place JCP would have concentrated, so we delved into what else they may have done.
We found something we didn’t expect. JC Penney has reformatted the URL scheme for its entire site! NONE of the URLs that we studied were now active. Every one of them now 301 redirects to the homepage. For a site to make such a massive change, where pretty much every category and product URL has been removed, is very unusual – unless there had been a complete re-platforming of the site.
Here is the breakdown of keywords based on the URL type that was ranking highest for the keyword:
Looking at the URL’s that JCPenney had ranking prior to the big lift, we saw an interestingly common pattern. Here is the breakdown of keywords based on the URL type that was ranking highest for the keyword:
|URL Type||URL Pattern||2/10/2011||5/9/2011||5/12/2011|
The rankings were dominated on the dates we studied by Product Category and Product URL’s of the following format . All of these now redirect to the home page.
Some examples of these URL’s are as follows:
|Category URLs||Product URLs|
There are two interesting things about this:
- Because these URLs still rank for the keywords, the change must be very recent and not yet completely updated in Google’s database
- All of these are redirected to the home page rather than an equivalent page, which is what should typically happen if this was part of a re-platforming project
What would prompt JCPenney to rebuild its site without adhering to the most basic of SEO rules to redirect to relevant pages ? Google’s cache allows us peek into what those pages looked like before they were wiped out.
Here is a sample of the old product URL:
and what the same product’s page looks like today
Also, here is one of an old category URL:
And the same category’s new page
A couple of striking differences between the old and the new pages caught our eye:
- While the core content above the fold seemed to be exactly the same, the footer links from the bottom of the page were completely gone! Look closely at the cached versions of the old page and you will see literally tens of very generic internal links. The new version of the pages are almost exactly the same sans those internal links in the footer
- JCPenney went from a fairly simple URL structure to a much more convoluted one with multiple query parameters. For Curio Cabinets category for instance, the URL changed from a simplehttp://www.jcpenney.com/products/Cg13344.jsp to this http://www.jcpenney.com/jcp/XGN.aspx?DeptID=70752&CatID=72423&SO=0…
What does JCPenney’s experience mean for your site?
As mentioned before, JCPenney has no doubt been working hard on cleaning up every possible thing on and off-site that may be construed as an attempt to “influence” rankings. The manner in which the URLs have been changed provides a strong signal that they may have been made to directly address the same issue.
So, is this the death knell for e-commerce sites using automatically generated, highly focused and optimized pages like the ones above? Or a precautionary measure by JCPenney to play it safe and getting rid of anything that may be even remotely construed as risky?
We will continue to observe JCPenney’s progress and compare it to other e-commerce sites. In the meantime, we would love to know what you think about our findings and whether we’ve drawn the right conclusions. Please let us know your thoughts in the comments below.