Backlinks from older existing blog posts = good?

[Jooky]

I noticed a couple sites advertising backlinks from old blog posts. Basically the posts are all already live, they just edit the current text to either:

1) add in your anchor+link (i.e. add a new sentence/paragraph that includes your anchor+link)
or
2) make existing text link out with your link (so no choice of anchor text, no new content added)


Does this work to help rankings? Wouldn't Google find it weird that an 8-month old post was edited to only add an outgoing link?

Would it help if technique 1 was used PLUS the "Last Updated" date was updated? ("Deserves Freshness" signaling)

 

[Darth]

I've always wondered the same thing, can't Google easily tell the link is suspect?

 

[ryandiscord]

I'm not sure how Google sees it but, it doesn't seem much different than someone snagging your image and months later you asking them to credit your site. Unless the site sells many of these links and creates an unusual pattern of it happening.

 

[Andg]

How is that any different from repurposing a car domain to dental website just to link out from it?

 

[Ryuzaki]

I'd say this boils down to "authority & trust."

Google can't have their algorithm assume that each case is manipulation. It's too resource intensive. Corrections, clarifications, returning to add sources and additional depth, giving credit for a photo, etc. There are plenty of reasons that this happens naturally, too.

And they aren't looking at this manually, so there's likely some threshold where they greatly reduce the chance of it being manipulation after they ignore it, likely based on some internal page rank and trust rank metrics... of both websites involved.

Then of course there's relevancy of the anchor text to the destination page, relevancy of the text surrounding the link, relevancy of the topic as a whole, etc.

My guess is they also compare the new version with the cached version and restrict the flow of page rank through edits that include only a link. But even if that's the case, I've seen it make incredible differences in rankings.

Hell, I could point you to a set of SERPs where a guy made an image about 15 years ago and it's spread all over the net, and he has Google Alerts or something going and all he does is email around asking for a link citation for its use. He has an unbelievable backlink profile where nearly all of it is this type of "after-the-fact" edit. His site is ranking more than 'just fine.'

 

[Jooky]

Thanks for your thoughts

Hell, I could point you to a set of SERPs where a guy made an image about 15 years ago and it's spread all over the net, and he has Google Alerts or something going and all he does is email around asking for a link citation for its use. He has an unbelievable backlink profile where nearly all of it is this type of "after-the-fact" edit. His site is ranking more than 'just fine.'

Sure, but those are fairly recent posts. My doubts are for the 5-year-old posts that get updated just to add a link. But like you mentioned, the algo may not be factoring in age + content difference yet...

I guess I'll just test these as there's very little chance of a manual penalty.

 

[GeorgeG]

But like you mentioned, the also may not be factoring in age

Yep. One good example is the bot does not rank the first version of an article but the most authoritative version.

E.g. if you publish a blog post and then a month later (after it is indexed) you decide to syndicate it to Inc, Google will rank that article on Inc and could even deindex your version (I've seen that happen.)

 

[turbin3]

Google can't have their algorithm assume that each case is manipulation. It's too resource intensive. Corrections, clarifications, returning to add sources and additional depth, giving credit for a photo, etc. There are plenty of reasons that this happens naturally, too.

This.

At the end of the day, it's algorithms, and algorithms = programming language. They can't know what's in your heart, only what can be parsed and understood at a code level. So there's nothing really esoteric about it. If they apply the algorithms indiscriminately, they'll "kill" a lot of innocent bystanders.

My guess is they also compare the new version with the cached version and restrict the flow of page rank through edits that include only a link. But even if that's the case, I've seen it make incredible differences in rankings.

And this.

It was clearly stated years ago by Google that they track the history of pages with each crawl, and they specifically monitor page changes. It stands to reason that, a minimal change like 1 link + 1 anchor text might equal a possible spam indicator. That would be easy from an algorithmic standpoint, with minimal risk to innocent bystanders. Combine that with other historical records, like similar behavior seen across an entire site, linking to sites of irrelevant topical themes, and maybe those possible indicators start to move past a threshold where it is clearly spam.

That being said, there can still be value in it. You just have to think creatively about how you can use "innocent bystanders" as a body shield, so the risk is too great for the algorithms to take their shot.

One example might be a link included along with an additional paragraph or several, added to an old post, all topically relevant, with context and other indicators (updated date) that the post is being refreshed with new content. That might be worth buying. Think about updates with new trends in a niche/industry, changes to old data that's now incorrect, adding additional resources or insights, etc.

If it was just a link and barely a sentence added, without seeing the specific opportunity, I would broadly say that's probably low value, but I could be wrong.

One thing is for sure, however. I wouldn't be "buying" over G's mail or other services, and I wouldn't take anyone seriously that is selling over their services as well. And if you're not sure if they're using G's mail with a custom domain, send them one from an anonymous/throwaway non-G email and see if the headers come back with G in them before you do business. I can't overstate the importance of this.

 

[Jooky]

One thing is for sure, however. I wouldn't be "buying" over G's mail or other services, and I wouldn't take anyone seriously that is selling over their services as well. And if you're not sure if they're using G's mail with a custom domain, send them one from an anonymous/throwaway non-G email and see if the headers come back with G in them before you do business. I can't overstate the importance of this.

Slightly off-topic, but can you elaborate on this? Are they checking for dollar signs + url patterns? Has this happened back on the "PBNs are dead" days? I know some PBN sellers send a pic of the urls to avoid tracking...

 

[Calamari]

I've done this. It works. But I would not buy a link like this. There's just too many factors at play.

 

[turbin3]

Slightly off-topic, but can you elaborate on this? Are they checking for dollar signs + url patterns? Has this happened back on the "PBNs are dead" days? I know some PBN sellers send a pic of the urls to avoid tracking...

I don't know exactly what they're doing or how, and I doubt anyone outside of the company can say for sure. We could talk about parsing strategies. We could talk about OCR and actually extracting text from images... However it might theoretically work is merely the extraneous mechanics of programming language. It's not really worth worrying about and trying to gauge it. That would be a no-ROI waste of time that wouldn't really lead anywhere.

The platform is designed entirely to monopolize and profit off of "free data". It stands to reason, then, that they track, analyze, and utilize every single piece of user-generated data on every single one of their tools and platforms. In fact, a number of court cases have surfaced evidence that demonstrates exactly that.

The only way to be sure...is to be sure by not using ANY of their services if engaging in that sort of stuff. That and getting to know EXACTLY who the seller is that you're dealing with, as opposed to some nameless and faceless forum name like most are.