Panda recovery with added internal links?

[Sgt_Kickaxe]

Disclaimer: it's not my site I'm discussing and I don't have access to server logs(or Google's crystal ball) but I can say that I've just witnessed a site make an apparent huge recovery and then some. I've also seen the changes made to get the new higher rankings which is what I'll mention here. There doesn't seem to be any new spike in backlinks.

The site, according to its owner, was hit rather harshly by Panda earlier this year(verious SEO tools agree) despite having original articles and not doing anything blackhat to speak of. After losing 50% of his traffic the owner added 75(!) links at the bottom of every page to other important pages. These 75 links are in addition to the 35 he added to the sitebar which already had 35. In total each page has 110 new internal links, 145 per page in total, and nothing else has changed.

The site now has more traffic than it did before being pandalized and, more importantly, is now top 3 for a good number of impressive keywords. Surely the increase in internal links can't be the reason?

[diberry]

If anyone can write about the effects that internal link changes (e.g. anchor text used, quantity) have made to their sites after Penguin, that would be great.

My Penguinized site is a mystery. The best guess several SEO people could come up with was that I had "too many" outbound links on some pages, so even though a human would recognize them as useful topical resources, the algo might see it as a link farm. I decreased a lot of those links (cut one page's links down by 80%, others 20-50%) and also increased the ratio of text to link.

No recovery at all.

[ken_b]

OK, I experimented with this idea a little, so I thought I'd share the results so far.

On June 17, 2012 I added 52 internal links to 51 primary pages and about about 900 pages one level down. All of those links point back to one or another of the primary pages.

So all 52 of the primary pages picked up over 900 internal links.

On June 17, of the 52 primary pages 17 ranked on page 1 for there primary localized query, (location+kw1kw2).

Since the the number of pages on page one has bounced around from 21 - 25 but seems to have more or less settled on 22 so far.

Probably too soon for these numbers to mean much, but I thought I'd share this now in light of the Panda attack on June 25.

.

[themaninthejar]

Maybe not directly relevant, but an observation on internal links:

I have a logo in the top left corner of every page which is a clickable link to the home page. Formerly I used my main keyphrase as the alt text. To reduce keyword densities I deleted the alt and left it blank.

After this I noticed in WMT that the internal links recognised for the home page started to drop off significantly - even though there's a text link to the homepage at the top of my side navigation (which is below the logo).

I filled the empty alt on the logos with the site's "brand-name" (primarily to increase brand name against keyword density). Now the internal link count for the homepage is increasing again.

On the face of it this looks like internal links have to pass some kind of quality test or they get discounted...? Which would isolate the target page and reduce its perceived importance within the site with a knock-on effect on ranking?