Rankings and search traffic fell off a cliff

[websearchseo]

Hi Moz community,

One of my clients has a beast of a website built in ASP.NET (which causes me problems cos I don't have much experience in that)

It is a job-site that aggregates job opportunities from other job-sites and provides a job matching service by email etc.

They used to have great presence on Google naturally for thousands of job searches.

Since Penguin and Penguin 2.0 (I think) their traffic has fallen off a cliff.

I have been doing some "off-page" experimentation, seeing if we can fix a lot of issues by re-sculpting their backlink profile (seeing as it was after penguin). but what I have found is that some pages respond to this off page work but some just do not at all, despite how we approach it, such as disavowing previous links building fresh new top quality content links with natural anchor text etc....

Which has lead me to the conclusion that the wider issue is on-page and potentially site structure. Unfortunately as it is ASP.NET I am not so comfortable diagnosing the issues.

I think also some issues will be related to dupe content etc.... but I would LOVE to get some input from my learned Moz colleagues.

The website is http://www.allthetopbananas.com/ - any tips on how to recover from this dramatic loss of traffic would be massively appreciated.

Kind regards

[websearchseo]

Hi Marcus,

Your help has been invaluable. Thank you so much for taking the time to do this for me.

Kind regards

Mark

[Marcus_Miller]

Hey Mark, given the Panda correlation I would look at content issues first and see what we can dig up there.

So, the 60 second SEO audit:

1. I grabbed a chunk of content from one page:

"Female PE Teacher, Permanent - Bath - Pay to Scale Full time role paid to scale depending on relevant teaching experience Required for September 2013 Full time as in accordance with the Teachers"

Run that through Google and what do we see.....

Lots of listings for that exact piece of content - approx 40 reported. But, we hit the third page and we see:

In order to show you the most relevant results, we have omitted some entries very similar to the 21 already displayed.
If you like, you can repeat the search with the omitted results included.

No sign of your clients site though!

So, we open it up and look at the omitted results and there are now over a hundred results and your clients site is in something like 95th position.

Conclusions?

The client site is basically a conduit to other sites like totaljobs.co.uk and re-purposes content available in several other locations. Therefore, they are being filtered as they are not the originator of the content and they bring nothing new to the table.

Pros & Cons

Pro - It's content so can be sorted easily enough by creating unique content.

Con - there seems to be a lot of it

---

Your client needs to do a content audit and create descriptions of the jobs that add some value and uniqueness. Given how many other folks there are pumping out these same descriptions there is likely some opportunity there but the old model is broken and they have to bring some value.

Caveat - this was the quickest of quick looks but certainly, it seems there are no real mysteries here so just dig in and do some deeper analysis. I imagine the site might give copyscape.com a heart attack!

Hope that helps!
Marcus

[websearchseo]

Hi Marcus,

Many thanks for your response. I have found the tools you recommended (Barracuda Panguin Tool and Link detox) extremely useful.

I have attached the barracuda results for allthetopbananas.com. It looks like Panda is the culprit.

Their link profile returned a result of "Low Risk" in Link Detox as expected, so I don't think Penguin is an issue.

If you have any further suggestions or direction from here, I would be most grateful. I am thinking that sites like this one, that have relied on basic on-page keyword optimisation and backlinks to attain their Google rankings in the past are now suffering because since Panda Google is putting more value on a wider set of signals such as social, authority, freshness, authorRank, mentions etc.. Would you concur with that?

I look forward to your reply.

kind regards

Mark

[Panguin Tool - Barracuda Digital-year.png](http://www.websearchseo.co.uk/dloads/Panguin Tool - Barracuda Digital-year.png) [Panguin Tool - Barracuda Digital.png](http://www.websearchseo.co.uk/dloads/Panguin Tool - Barracuda Digital.png)

[Marcus_Miller]

Penguin can and often does have page specific problems. In fact, most often it seems the homepages of many sites are most affected and seemingly Penguin 2.0 does a better job of targeting inner pages.

I have worked with several companies who are still struggling to rank their homepage after Penguin yet can generate traffic to their inner pages.

Firstly, I would not jump to assume there are lots of problems with the site, perform an audit if you are unsure but don't make this as an assumption. Use known penalty dates to review traffic and anoint the analytics so you can get a clear picture of which updates caused the traffic drops for a quick and easy way to get more intel.

More often than not I am now advising people to just start over on a new domain and redirecting the old (not with a 301) to the new. This is painful, and many folks are holding onto domains but if clean up is not working and is not practical to continue down this route then a clean domain gives you a fresh start.

I have just had a quick look at the backlink profile in OSE and it does not look like your standard penguinised site though. Are you sure about the cause of the loss of traffic? Do the drops coincide with known penguin dates?

Really, the analytics should be your first port of call here and tools like this will help you easily map algorithm change dates to traffic changes: http://www.barracuda-digital.co.uk/panguin-tool/

Then, investigate further from there on in. If links do seem to be an issue link detox is a quick and easy way to get a good grasp on what kind of links your client has.

Ultimately, it needs investigation but start with analytics, form your hypothesis and try to prove it and then feed back here.

Hope that helps

Marcus