|Getting Past Panda I - Third Attempt|
My third attempt at getting away from the original Panda. Thanks to some extremely useful suggestions, I have finished a four week review and changes of the site in question.
Remove or condense pages written in the 2010 period (some older pages as well). Most were written for an expansion of subject matter, while some were in reaction to Ehow sticking to the top of the SERPS for many subjects I am very familiar with professionally. See, unlike most Ehow writers, some of us actually have the real life experience to expound on a subject.
Pages that had poor user numbers (time on page less than 60 seconds) were either removed completely or consolidated to other pages within the site that are of context. Nearly all of these pages were also only visible past page three of the Google SERPS. This resulted in a reduction of 27% which now shows total pages on site at 276.
Improved navigation for both Google (better understanding of the site overall) and the visitor.
Inverted L navigation, which I thought was a brilliant suggestion and I thought visitor numbers would show great improvement. Previously the site had left side navigation where only 27% was visible above the fold (1270 x 768). Adding in related page links at the end of each subject page and that number was more like 20- 24%.
All page links are now visible above the fold. Average visit duration (AVD) is showing an overall 26% increase, but page views per visitors (PVV) has actually declined 27%. This really has me baffled. I cannot figure out why. The only reason I can come up with is a decent number actually made it to the bottom of these pages previously.
Solution? Insert anchor point “see related links” at the bottom of every page? This would direct visitors to the top of the page, or in full view of the current related page links area.
Lousy Numbers: Landing Pages
Bounce rate increasing - + 12%
PVV landing page - 17%
AVD - 19%
Very Positive Numbers: Site Content/Overview/All Pages
Top 10 pages on the site are showing AVD increases form 30 to 200%. Nine increases and one basically flat.
I am crossing my fingers and feet hoping that Google will reward my efforts and return the site to a meaningful position in the SERPS. I truly believe the site now rises above the competition. Unfortunately the competition since Panda has become big brands. I’ll be happy with a small piece of the pie nowadays instead of crumbs:) Besides, I have been working on plans that do not depend on survival from free web traffic.
Meaningful to the point I can again feel proud of something unique.
|Average visit duration (AVD) is showing an overall 26% increase, but page views per visitors (PVV) has actually declined 27%. |
Hmmm - maybe your site's "information scent" was strong enough to keep people poking around, but the old navigation had them checking out too many false leads?
Sorry Tedster, I don't get the false leads.
False lead = "Maybe this link has the information I'm looking for." - "Whoops, no it isn't. Maybe I'll try another link here first instead of going back to Google right away."
Just one scenario I thought of that could match your shift in stats.
I'm very interested in the results of your shift to an inverted L navigation. I know it hasn't been in style recently, but sites where I use it still seem to thrive and sites with other trendier approaches often seem to struggle.
Not that such a change would DIRECTLY affect a Panda demotion, but it certainly might do so indirectly.
[edited by: tedster at 11:45 pm (utc) on Oct 7, 2012]
I’m now thinking it may have to do with an increased number of 404 pages from Bing and Yahoo visitors. In looking through some recent discussions about using 301’s it seemed to be the best option to let it serve file not found. A good number of those pages ranked very well before my changes (not Google) but they were weak and fugly looking. I always detested pages that showed a large amount white space before the end of the left side main navigation...if that makes any sense.
Lostone, does your site have any sections that get a lot of google traffic, are of very little value to you (you dont earn much from them) and the user metrics are very different to the site average?
For example, if your site exit rate is 60%, do you have high traffic sections with a 30% exit rate (or vice versa).
A number of sections that get reasonable traffic, but I never looked at earnings from them. I also never compared value in the sense of money either. It's never been about the money but more about creating a source that actually sees it from all angles. Not commercial or biased either.
LostOne, I followed your link from another thread. I want to make sure I understand what you're seeing. It sounds like your post-Panda INBOUND traffic has held steady, but since you made these changes visitors just aren't viewing as many pages during their visits, even though they're staying longer.
If that's the case, I do think Tedster's theory could make sense - before, they were bouncing around to pages that weren't really what they wanted, and now navigation is much more clear so that's not happening. If so, that's mixed news - obviously going to hurt pageviews, but should build more visitor loyalty.
Your solution of putting "related links" at the bottom of the page text should help - did you try that? Another option would be to cross-link related pages from within your text.
Don't forget the spam detection patent, which could delay any benefits from changes you've made: [webmasterworld.com ] It can impose a time delay on the rankings reflecting your improvements.
Well they say persistance wins in the end. Churchill once said "Success is stumbling from failure to failure with no loss of enthusiasm.”
I never saw that discussion dib. I have not touched anything since all the changes were made in early October. I don't plan to either. Just have to ride it out. Prior to all this I did have related links at the bottom.
Time on site has crashed as well, or down 35% The first few days it looked good then it tanked. I have some backup plans...one needs to these days. One includes "if you can't bet 'em, join 'em." That refers more to getting together with a big brand in my industry. I can do quite a bit for them with my ideas and I don't think they're too big so as I have to wade through dozens of departments to get something going. Only time will tell as this is a new step. At least they know my background and I'm not a nobody in the business. Too early to tell.
Like I've said previously, it's never been about the money, but just getting back in the loop sort of speak, and being able to create without thinking about the Google gavel coming down at any moment.