Effects of losing valuable links

[bbarbeau]

Has anyone had any experience with losing high profile links?

My company previously partnered with a high profile domain in our vertical to provide them with web services on their site. As a result of the tool we provided them, we had a "powered by ______" image link along with some anchor text pointing back to our homepage on every page that this tool lived on (probably several thousand pages of their site, with PR varying from 0 to 6, though the vast majority were 0).

The anchor text in those links wasn't exact-match for high-volume keywords or anything like that. If you put the entire anchortext in a keyword tool, it would return 0 for search volume. (The anchortext is our name and some decent keywords in them, but nothing head-of-search in our vertical.)

Anyhow, the partnership is concluding, which means these links are not long for this earth.

I wasn't here when the links went up, so I can't say what kind of boost it gave us at the beginning of the partnership. But I'm here when the links are going to come down and am just wondering how significant this change can be for our rankings.

Anyone had similar experiences, and if so what kind of impact did it have?

[martinibuster]

I can't answer your question from personal experience. However in my opinion it's possible your rankings may be supported by other factors. Here is why I think so:

1. "Powered by" links have been targeted for depreciation for almost ten years. They have steadily been losing efficacy (for the purposes of ranking) ever since.

2. The location of the link within the web pages might be cause for a depreciation of the usefulness of the link for ranking purposes.

3. The fact that the link is run of site does not make it more powerful either. So the usefulness of the link may have been overstated to begin with.

4. Words such as "powered by" could be cause for diminution of usefulness of the links for ranking purposes.

All of the above are reasons to be suspect about the usefulness of the links. Let us know if the removal of those links does indeed cause a change in ranking. It's possible there won't be much of a change.

[jakebohall]

In my experience in working with companies using this tactic with template and widget bait, it can go either way. In many instances the links were mostly devalued and the impacts were negligible.

This could be your case due to the fact these are sitewide and easily algorithmically identifiable as Martinibuster pointed out... and we are talking about a single domain.

Since it sounds like this change is inevitable, I would suggest focusing your effort on finding a few authority links to replace the one you will be losing. I'm looking forward to hearing how it plays out for you.

[bbarbeau]

Well, I'm optimistic hearing from you guys that losing these links may not cause us a rankings apocalypse, but I've also got to admit that I'm a bit disheartened that a strategy like this wouldn't have been as powerful in the first place as I thought. I didn't know that "powered by" links had been targeted by Google, and for so long, too.

To be clear: the words "powered by ________" are contained in the image, so the text "powered by" doesn't actually appear in the source. But the vast and consistent placement across thousands of pages... I'm sure their algorithms were able to figure it out. If it makes a difference, the link was fairly visible on the page--above the fold, not in the footer or sidebar or anything.

Anyhow, they'll be going away within the next 60-90 days, so I'll try to follow up with a post about any effects/non-effects that this has. If anyone comes across this thread several months from now and I haven't posted results, post a reply here and I'll update you all with the results. But I'll also try to remember to update the thread in the weeks following the links coming down.

[martinibuster]

I didn't know that "powered by" links had been targeted by Google, and for so long, too.

I heard it straight from Marissa Mayer at SES San Jose in 2003 or 2004 in a session about search crawlers. She said they would depreciate powered by footers from irrelevant sites. I raised my hand and asked her that specific question. Her answer was coming more from the point of relevance than the position of the link on the page. However the point is that was the first time anyone from Google had stated that this kind of link evaluation was happening and it was a bombshell revelation at the time. That was the first step in the evolution of the algorithm toward depreciating links like you described. There were more changes over the course of the next years in the way links are handled.

Later on Google introduced depreciation of links based on location within the web page. There already was a crude version of that in the original search engine, where content at the top of the page was deemed more important than content near the bottom, as well as matching title tags to anchor text etc. The idea of all these ways of treating links was to identify relevance but at Google was increasingly gamed the emphasis started tilting more toward identifying attempts to manipulate the rankings, figuring out how to separate real citations from those that are not. It's all been evolving year after year.

[bbarbeau]

Thanks to martinibuster for asking me to update this.

So we lost these links on 6/1, which was a little over a week after the original post. I'd estimated 60-90 days initially until the links went away. We thought we'd have that much time to begin winding down the site, but they wanted to transition at the end of the month so we obliged.

At that time we were ranking #7 for the head of search term which was included in the anchor text but not as an exact match.

Here's a timeline of what's happened since. (My rank checks are reported weekly):

5/27 - homepage ranked #7 for query
6/1 - site goes down, 301ing to our home page
6/2 - homepage ranked #8 for query
6/9 - homepage ranked #6 for query
6/16 - homepage ranked #8 for query
6/23 - homepage disappears from serps (just for this query); interior page ranks #7 for query
6/30 - interior page ranks #7 for query
7/7 - interior page ranks #10 for query; homepage reappears at #15 as well
7/14 - interior page ranks #9 for query; homepage disappears from serps (just for this query)
7/21 - interior page ranks #11; homepage reappears at #12
7/28 - interior page ranks #10; homepage remains at #12
8/4 - homepage ranks #11; interior page ranks #12
8/11 - interior page ranks #10; homepage ranks #14


Looking at keyword visibility in general for many terms I track, nearly 200 have dropped from page 1 to page 2 since the beginning of June.

However, I think it's important to note that the SEO woes I've faced this year are not only limited to losing links from this one high profile site.

This thread [webmasterworld.com ] details issues I've faced with our URLs being manipulated by Google such that the URLs we use in internal site linking, sitemaps, PPC, emails, natural and acquired inbound links, etc. are being discarded by Google in favor of URLs that they create (this is based on poor 404 behavior on our site, along with a lack of canonicals).

Also, we had been on the slide most of the year for this query anyways, beginning the year at #2, then slipping to the bottom of page 1 where we are now.

So to attribute all of this to the loss of these links is not accurate and only part of the picture.

However, I do believe those links going away has had an impact, not only in rankings, but in what is ranking.

The internal page that Google now seems to prefer for this particular head of search query always hovered in the 90s in terms of its ranking for the same term. Just prior to Penguin 2.0, it had risen to the high 40s in rank. (No optimization was done to effect that change.)

Penguin 2.0 launched and it dropped from the serps as part of 2.0's redundant domains tweak until 3 weeks after we lose these links, where it shows up again as the new most-relevant page in Google's eyes.

[jakebohall]

Thank you for the update. When you look in Webmaster tools, do you see any manual actions against your site?

While I would have to think that losing these links would have some impact, it seems like there are many factors at play.

I would definitely recommend a solid review of your backlink profile before making any decisions for new link development, removals, etc..