Google is apparently getting links wrong from time to time. By wrong, we mean giving webmasters example links (in unnatural link warning messaging) that are actually legitimate, natural links.

It’s possible that the instances discussed here are extremely rare cases, but how do we know? It’s concerning that we’re seeing these stories appear so close together. Do you think this is an issue that is happening a lot? Let us know in the comments.

A couple weeks ago, a forum thread received some attention when a webmaster claimed that this happened to him. Eventually Google responded, not quite admitting a mistake, but not denying it either. A Googler told him:

Thanks for your feedback on the example links sent to you in your reconsideration request. We’ll use your comments to improve the messaging and example links that we send.

If you believe that your site no longer violates Google Webmaster Guidelines, you can file a new reconsideration request, and we’ll re-evaluate your site for reconsideration.

Like I said, not exactly an admission of guilt, but it pretty much sounds like they’re acknowledging the merit of the guy’s claims, and keeping these findings in mind to avoid making similar mistakes in the future. That’s just one interpretation, so do with that what you will.

Now, however, we see a Googler clearly admitting a mistake when it provided a webmaster with one of those example URLs for a DMOZ link. Barry Schwartz at Search Engine Roundtable, who pointed out the other thread initially, managed to find this Google+ discussion from even earlier.

Dave Cain shared the message he got from Google, which included the DMOZ link, and tagged Google’s Matt Cutts and John Mueller in the post. Mueller responded, saying, “That particular DMOZ/ODP link-example sounds like a mistake on our side.”

“Keep in mind that these are just examples — fixing (or knowing that you can ignore) one of them, doesn’t mean that there’s nothing else to fix,” he added. “With that in mind, I’d still double-check to see if there are other issues before submitting a reconsideration request, so that you’re a bit more certain that things are really resolved (otherwise it’s just a bit of time wasted with back & forth).”

Cain asked, ” Because of the types of links that were flagged in the RR response (which appear to be false negatives . i.e DMOZ/ODP), would it be safe to assume that the disavow file wasn’t processed with the RR?”

Mueller said that “usually” submitting both at the same time is no problem, adding, “So I imagine it’s more a matter of the webspam team expecting more.”

It’s a good thing Mueller did suggest that Google made a mistake, given the link in question was from DMOZ. There are a lot of links in DMOZ, and that could have created another wave in the ocean of link hysteria. Directories in general have already seen a great deal of requests for link removals.

Here’s a video from a couple summers ago with Cutts giving an update on how Google thinks about DMOZ.

Cutts, of the webpspam team, did not weigh in on Cain’s conversation with Mueller (which took place on August 20th).

Mistakes happen, and Google is not above that. However, seeing one case where Google is openly admitting a mistake so close to another case where it looks like they probably also made a mistake is somewhat troubling, considering all the hysteria we’ve seen over linking over the past year and a half.

It does make you wonder how often it’s happening.

Update: Just got a tweet from Cutts on the matter:

Do you think these are most likely rarities, or do you believe Google is getting things wrong often? Share your thoughts.

Image: Google