Domino Effect of Mass Link Changes

[jakegotmail]

If one site abruptly made significant changes to their link profile, could it trigger a ripple effect that could be felt across many sites?

We recently began making changes to remove some old, bad links that accounted for a unacceptably large percentage of our link profile -- on the order of 65%+ on a profile of nearly 2 million links.

While the change is just recently starting to show up in WMT -- only about 10-15% of them are gone so far -- we had seen some small positive gains in ranks and traffic, which was encouraging. While those gains have leveled off, I am now noticing that a number of other sites in our profile appear to be targeted by Google and have dropped in PR -- PR5's and 6's dropping to 2's and 3's, and 3's and 4's dropping to 0 -- including a couple sites that stopped linking to us over a year ago.

While I can't be absolutely sure on the timing of the PR drop, I do know that the change did not occur prior to the links being removed about a month ago, and I'd have a hard time believing that the timing is coincidental.

Is there any precedent for this behavior?

I've definitely seen it work in the opposite direction, from the top down, but not from the bottom up.

[goodroi]

If one of your sites triggered a manual review, it would be easy for Google to identify the rest of your site network. I have seen one bad website lead to many bad websites being impacted.

[jakegotmail]

I could see that if our site had been hit as well, but it hasn't. Our traffic, rankings, and PR remain the same as before, and there is no indication of a manual review (no notices in WMT at least).

And I wouldn't call this a network by any means. It's just old school link building done in a not so great manner.

My curiosity stems from the fact that we're working to correct this, and now we see other sites getting knocked with little to no rhyme or reason. There's been no change to the handful of sites that had the most links on them, while others showed a drop in PR -- but as I said, that set included sites that we haven't worked with in over a year.

There has to be something else at play here, right?

[jakegotmail]

Does anyone else have any random ideas on the subject or related?

This pattern doesn't seem natural. I'd almost feel more comfortable if our site had been hit -- at least that would have made more sense.

[Sgt_Kickaxe]

There hasn't been a PR update in a long time so I'm not sure what change your referring to but to answer your question...

The internet is fluid, a change in one place can and does lead to changes in other places. Since Google places such heavy importance on links it is conceivable that you removing a link to a page with very low visibility, and with only a long tail keyword rank, might cause it to lose some positioning. That's highly unlikely however, there are many mechanisms in place that limit and delay positive change as well as many that dilute the impact any single site can have on the net.

Have you made sure you're not seeing personalized results for the sites you say are falling?

[jakegotmail]

I'm referring to a drop in PR in multiple Chrome extensions that have always been reliable in their reporting. While they claim that there have been no PR updates in some time, there are obviously still some adjustments being made in some fashion -- manual or otherwise.

This is strictly a PR question. Nothing about the issue I'm describing is in any way (yet, after possibly a month) related to a drop in rank for any pages, for any terms -- nothing that I could point to as being statistically significant that is.

Some terms have risen, some have remained static, and some have dropped -- but nothing drastically.

There is no correlation between the terms that have moved and whether or not they are found on sites that appear to have been targeted or not. Likewise, there is no connection between terms movement and whether or not they are found on sites where we removed our links.

I have no knowledge of whether or not these sites have seen a drop in traffic or keyword rank, but I can assume with the observed drop in PR that they have seen these decrease.

In this instance, there were a handful of sites that we identified that we wanted to clean up -- these are the ones mentioned in my first post -- none of which dropped in PR, and no terms we were using in anchors showed any real movement.

Of the group of sites that we observed that did drop in PR, 80% of those we were still working with, but 20% we haven't worked with in over a year (no links still found on their site), and likewise, no terms we are or were using showed any real movement.

To date, I'm unable to identify any common thread in the way of changes in the last month between all of these sites, other than our working with them. That isn't to say that most aren't still doing the same things with links that they were/are doing with us.

This is where the usual top-down flow would make sense; if these sites got hit or made some massive change, and that affected all of the sites it was linking out to. That's the norm. But this is almost flowing upstream -- we're not all linking back to these site, but that because we made massive changes at the bottom, we somehow flagged a review that flowed backwards, but only affected sites sporadically, and didn't affect us negatively.

As I said, the timing of the observed drop in PR aligns too well with our massive housecleaning for me to think that it isn't somehow related to our site and our link profile specifically. If I identify some other common thread, I'll definitely pass that along, but for now, I'm looking for any other experiences similar to my own.

[jakegotmail]

Would love to hear any more input on this subject -- previous or current experiences

[lucy24]

Short version: You took action to improve a group of sites-- in a way that should apply both to humans and to search engines-- and one result was that some other sites ended up doing worse in searches? Did I get that right?

[jakegotmail]

It would appear that way -- although I'm basing that assumption strictly from the fact that I am observing their PR drop. I have no data to say that they have dropped for any keywords.