Editor's note: Whew! This one generated a fair bit of controversy in the SEO community. Be sure to check out the comments below the post for a holistic picture of the debate at hand. Carry on!
Penguin is back at the forefront of many marketers' minds now that the third iteration of the algorithm update has been released, and a rumor has begun circulating that you can weasel your way out of a Penguin penalty by simply submitting a disavow file. In today's Whiteboard Friday, Josh Bachynski breaks down that argument and starts a realistic discussion to find the answer. While we (and, as you'll see, Josh) don't have definitive answers, we hope you'll join in with your thoughts in the comments!
For reference, here's a still of this week's whiteboard!
Video transcription
Hi. Welcome to Whiteboard Friday, and I'm your guest host for this week, Josh Bachynski. This week I'd like to talk about whether or not it is true that you can disavow your way out of the dreaded Penguin algorithm.
So there is a hypothesis going around the SEO community that it is possible that you can just use a disavow file to get out of Penguin. Now, for those of you who don't know, the disavow file is a feature that Google implemented a couple years ago where you can upload your spammy links into a file, very similar to robots.txt, and they will apparently remove those links out of your link graph or have them not count against you or something along those lines.
However, the hypothesis is -- and Google has confirmed this both by John Mueller and Matt Cutts -- that apparently if someone sends you bad links or you make bad links, however the link showed up, you can just disavow those links, you can put them in your disavow file, and this will help you get out of the Penguin algorithm.
So this is the hypothesis, and not only does Google claim this is the case, but many SEOs claim this is the case as well. In fact, they go so far as to claim to have succeeded doing this for themselves or for clients, that they have just taken links and put them in the disavow file and those clients, on a Penguin refresh, have been saved from the terrible Penguin.
It would be a large problem if this was not the case, because, as I mentioned, it is possible for people to simply buy a Fiverr blast -- I don't want to list off too many options to give you negative SEO ideas -- but you could imagine scenarios where it's pretty easy to build these spammy links pointing at sites and possibly get Google to notice them and then to milk those sites when Penguin 3 comes around or Penguin 4, the next iteration of Penguin.
It would be very good if the disavow file worked. Personally, I'd like the disavow file to work if I could prove that it did. It is a problem, in a lot of ways, that it's not, which I'll get to in a second.
However, if this is true, then there should be no recovery with link loss. For example, if this is true, that you can just disavow your way out of Penguin, then we should be able to find sites that escape on the Penguin date, but have deleted no links or we can tell have had no link loss whatsoever. That way, we can know that it was just a disavow file and not some combination of either deleting links on the disavow file or something else entirely. So if that is the case, that this hypothesis is true, then we can use a scientific method to determine that we should be able to find exemplars of the hypothesis.
On the last Penguin 3, when it was released October 18th, for those people who claimed that they recovered and claim they did it only from a disavow file, I asked them to send me examples. I said, "Fine. Send me your URL, and I'd like to check it." I tested over 12 sites altogether who claimed to have both recovered from Penguin on that date and to only use the disavow file or claimed to only use disavow file to do so.
However, I found something rather striking, that every single one that I checked, they all had link loss. In either Majestic or Ahrefs or using the Moz tools, I found that they all had links that they lost a few months prior to the release of Penguin.
Now whether they deleted the links and just lied to me, or whether they forgot they deleted the links, or whether the links just dropped off the link graph because, of course, web pages on the Internet change. For all we know, these could have been just scraper sites scraping them, giving them links that they didn't even want, and those sites just disappear. However the links were lost, the links were lost.
So, what does that tell us? Well, unfortunately, it tells us that I cannot confirm the hypothesis. After 12 tries, the hypothesis that you can just use the disavow file to escape Penguin, I was not able to confirm that hypothesis. The examples, the evidence that people sent me trying to prove this hypothesis proved to be false. So I say myth busted or at the very least myth not confirmed. I was not able to confirm it after 12 plus tries to do so.
At best, all I can say after doing the testing and, of course, I just want to add this note in now, if anybody out there, anybody seeing this video claims to have recovered from Penguin and just done it solely from the disavow file and they didn't delete any links and they didn't lose any links, please, by all means, send it to me, because as I said, it'd be lovely. It'd be wonderful if that's the way it worked, because then if someone is sending you a negative SEO attack, all you have to do is watch your backlinks on a daily basis and throw in there any ones that seems suspect.
But as I said, I could not confirm that's the way it works. At best, all I can confirm is that deletion of links or loss of links still apparently has to be required in some way, and then two, this experiment, of course, has absolutely no bearing whatsoever on the manual penalty process, which I won't even get into, which the disavow file may or may not help with. I'm not talking about that for this Whiteboard Friday.
The question then you'll ask me is, "Josh, why, why, why, oh why, do people perpetuate this myth?" Well, I'm afraid there is a number of plausible reasons why they might perpetuate this myth, both Google and other SEOs. One is because it's easy. An SEO who knows half of what they're doing can get a list of links and put them in a disavow file and give them to a client in about five minutes to upload. In fact, there are programs that will do it for you very quickly. Are they selling snake oil? I don't know, but I could not prove that the disavow file helped in any way, shape, or form for trying to get out of Penguin.
Two, there is another reason why Google might possibly -- I'm just putting it out there for your consideration -- perpetuate the myth -- as far as I can tell it's a myth -- that the disavow file will help you escape from Penguin is because you're feeding their machine learning. Every link you put in there, it's entirely possible they can run through their algorithms, which Matt Cutts has admitted, at SMX Advanced 2013, they might just think of doing at some point in future, so they can tell what these badder spammy links are.
And finally, propaganda. People are very afraid of negative SEO, with good reason. Whether or not it works or not, it definitely is a scary concept, and so it would be very reassuring for Google to tell people that, "Hey, we have this nifty disavow file. So if you get scared, if you see some suspect links pointing to you, all you have to do is put them in your disavow file, and you don't have to worry about it at all whatsoever."
However, I'd love that to be true, but I was not able to prove that being the case. So I'm going to say that I think the myth is busted. If anybody has any counter evidence to send to me, by all means I am all ears to look at it. All I need to do is plug it into a Majestic SEO or Ahrefs and see if there are any deleted links before the last Penguin release and say, "No, you lost links, and so we cannot say that it is the disavow alone."
To confirm that hypothesis, I would have to see no links lost in Majestic and no links lost in Ahrefs whatsoever, and, of course, I'd have to see an uptick on a declared Penguin date for me to say, "Well, jeez, the evidence looks like they have released on Penguin, and they had no deleted links." Then I'll take your word for it that you submitted a disavow file, because, of course, I can't see that. Only the site owner can see that, or you can give me your login whatever. You can trust me.
Until that time, I'm saying the myth is busted. The disavow file alone does not help you escape from Penguin, maybe in combination with deleting links, I'm not sure. I'm saying the disavow file is, unfortunately, the opiate of the masses. It is a safe myth we believe in because it makes us feel warm and snuggly at night. But I'm afraid that, after scientific testing, I cannot prove that that is case.
I've come away from that with two more suggestions that I would recommend. One, I would stop paying for it. I would stop buying it. I would stop paying people to simply make you a disavow file and upload it. I would tell SEOs to stop selling that as a service alone. Of course, in conjunction with other services, fine. But that as a tactic alone, that's not going to do anything at all, because the evidence, so far that I've seen, doesn't suggest that it will.
Furthermore, a more general point, it might be a good idea to think about stop selling and stop buying from-the-hip SEO, where SEOs are selling services based merely on hearsay and as much as we can, in our industry, triangle more for science based SEO or data based SEO.
If anyone recommends any service to you or any suggestion or any SEO tactic to you, the first question you should ask is, "Where did you come by this information? Do you have any data to prove that this is a good thing to do?"
That is my Whiteboard Friday for this week. If you have any questions at all or you want to e-mail me, yell at me, contradict me by all means, or please send me more sites I can test that may have sites that didn't delete any links, but did see an uptick on Penguin. By all means, join in the comments below, or e-mail me at [email protected] with that or any other questions. With that, I bid you adieu, and we'll see you again next time. Bye-bye.
We have (Angular/Remove'em/formerly Virante) always recommended removal along side disavow. It may very well be possible to succeed with just disavow for now, who knows, but Matt Cutts made it clear...
"Well, first and foremost, the main thing that we recommend is getting those links actually removed from the web. If a random person looks up links to your site, you don't want to see a bunch of spammy links. It's good to clean up for other purposes. You don't want them to jump to conclusions. It's also good so that other search engines won't see spammy links to your site as well. We do recommend that you write to people, try to get the link taken down, and basically get as many of those links as possible."
Or again, on a different occasion, he states...
"The other thing that we see is sometimes people think that disavow is the be all end all, the panacea that's going to cure all their ills. Yet we do want, if you've been doing some bad seo and you're trying to cure it, in an ideal world you would actually clean up as many links as you can off the actual web. That's just a really helpful way for us to see, when you're doing a reconsideration request, that you're putting in the effort to try to make sure that things have have been corrected, cleaned up, and not going to happen again"
So what we have is a direct statement from Google on how to properly handle a link penalty. Even if the disavow worked alone, Google might choose to stop respecting disavows that do not include link removal in the future. All of this is speculation. The only thing we know for certain is that Google asks us to do both. So we do both.
If Google stop respecting disavow, then there'll be no cure for negative seo. Hundreds or thousands spammy websites with no contact will make Your website serps lowered and there could be no cure for it (as simple as disavow). So Imo Google can't stop respecting disavows and must follow it. And still disavow works fine with penalties but people expect miracles after hit by penguin and say "it doesn't work!".
I think the problem is that after people disavow, they expect to get the penalty lifted and get back to the initial rankings. But people do forget that they basically "delete" links from google, so in order to get back their rankings they have to build more links.
Hey Russ! Thanks for commenting! I know what Google says. John Mueller has been saying "use the disavow tool" for over a year. That is not the point :-) The point is trying to verify his statements! And also, more importantly, verify with evidence the statements of other SEOs who claim that a pure disavow will work :-) NO ONE has been able to send me a simple, clear example of a pure disavow occurring. I am at 50+ attempts of theirs now. Thus they cannot say or sell that a pure disavow works. Neither can we, John's admittedly biased statements notwithstanding. As I speculated in my video, there are "alternative" reasons why both Google and SEOs want to perpetuate the myth that a pure disavow can work...
Matt is referring to a manual penalty, not an algorithmic one. Article is referring to the penguin algorithmic penalty
The problem with this WBF is that it is pure conjecture. Josh has strong opinions about Penguin. I have strong opinions about Penguin and some of those differ from Josh's. But, it's really a tough tough thing to prove any Penguin theories. I respect having discussions like this though as it's helpful to get the thoughts from other people who are dealing with a lot of Penguin hit sites. But, I think we have to be careful to label this WBF as "theory" and not fact.
When Josh was looking at whether a site recovered from Penguin due to a pure disavow he was saying that it was not a pure disavow if the site had evidence of link loss. The problem is, that it would be very hard to find a site that had previously had unnatural links built and had NOT seen evidence of link loss. For example, in the past, if an SEO was building spammy links on spammydirectory.com and spammyarticlesite.com, a good number of those sites will have gone offline and would be considered as lost links even though the site did not do any link removal. So, I could look at a site who had done a pure disavow (no active link removal) and I'm still going to see evidence of links being lost.
I checked on the ahrefs report of a site where the site owner had recently approached me to help them do some work. The site owner said in their email that they had not done any link removal, they had simply disavowed. (They were looking to have me review their disavow file.) And look, here is their ahrefs link report which shows extensive link loss:
https://www.hiswebmarketing.com/linkedimages/link-loss.png
So, in Josh's experiments this would not classify as a pure disavow, but really it was.
And here is another site. This is one that has never had a single link manually removed. Yet, the chart shows link loss:
https://www.hiswebmarketing.com/linkedimages/link-loss2.png
As such, I don't think you can say, "Hey! This was not a pure disavow Penguin recovery" just because you're seeing evidence of link loss.
So, can you recover a Penguin hit site just by disavowing? Here are my recommendations:
1. Remove any link that you can easily remove. (Examples would be keyword anchored urls in the footers of clients' sites, directory links where you have the login, paid links where you can easily get the link removed etc.)
2. Disavow the rest. Make sure you do so on a domain level.
If it is an algorithmic hit (i.e. Penguin) I don't recommend doing a large scale email outreach for link removal. The reason why I don't is because in most cases the success rate is low. It depends on the niche, but for manual penalties when I have done thorough link removal campaigns (and I have done a LOT of these), the success rate is somewhere between 5% to 20%. So, in order for me to do a link removal campaign for a Penguin hit site I'm going to be charging a very large fee and the ROI is low.
I thought it was a little ironic that Josh was calling the people who recommend just disavowing (as opposed to doing extensive link removal) as snake oil salesman. What I have seen is that some companies will say, "We can disavow for $xxxx" or we can do a thorough link removal campaign for $xxxxx". In fact I've even had some SEO's tell me that they are not sure whether link removal is necessary, but they're going to keep recommending it because it is a big money maker for them!
I wrote a fairly extensive article on my thoughts (and what Google says) on disavowing vs removing links for Penguin recovery here: https://searchenginewatch.com/sew/how-to/2374406/penguin-recovery-should-you-be-removing-links-or-just-disavowing
I will hopefully be publishing an article on my findings soon. I've got a few sites that I feel recovered or saw significant improvement in the eyes of Penguin just by disavowing. The problem is though that it is almost impossible to say that a pure disavow is responsible for a site's recovery. In most cases, we also fixed other issues on the sites. Or, in some cases, we did a pure disavow and the site improved from position #8 to position #1 or #2 when Penguin refreshed. You could argue that perhaps Penguin caused their competitors to drop. Or, if a small site recovered from position #99 to position #8 there's not going to be much change on analytics to prove a recovery.
There is still a lot that we don't know and can't prove about Penguin!
"So, in Josh's experiments this would not classify as a pure disavow, but really it was." You believe this on what...? Faith? Magic? Sorry Marie you just don't understand the science here It's not my theory. it's yours. I am just pointing it out you haven't proven it yet. It's not my fault you haven't done so, or it may or may not be hard to do so (another assumption) I am not the one insisting the disavow works beyond what the evidence supports You can link to your articles all you like, it doesn't change the fact you haven't proven it yet :-(
it is impossible to prove as lower quality sites dissapear quicker. Further, the business model for these sites were based on links, that model no longer works. These sites are working with their clients on a daily basis to remove past links and have little to no demand for new ones. Additionally, Google has gone after hundreds of sites for allowing others to guest post, thus forcing thousands of links to be removed and encouraging many others to change their outbound link policies. Google has not penalized a couple of sites in a vacuum, penguin has disrupted the business model of sites with outbound links. Prior to Penguin, the risk of getting caught was low so these sites had plenty of customers willing to pay for links or give content for a guest post.
I have personally submitted well over 900 disavow files & monitored the results. I can definitively say that this *does not* happen. You will not get "recovery" from Penguin from losing/disavowing links.
On the other hand, I've submitted disavows, done no link building and seen recoveries. I examined each of these cases VERY closely and found that in all of them there were new links - just a lag effect on those links being found. I would suggest lagging link discovery would account for any rumours that this is possible.
(Also, if you do a disavow AND a penalty recovery, you COULD see a recovery without new links based on the reconsideration request. So disavow + 0 links could = recovery ... but only with another step in the middle.
It's pretty clear the Disavow submission and Penguin recovery process is deeper than just those two actions. Recovery is on a site-by-site basis, for sure. Good point.
Especially if "bad links" are 80% of total links...
@ Matt-Antonino, I believe your statement regarding the reconsideration request may be a very helpful point for a clear understanding of the processes dealing with the disavow topic
@ Matt-Antonino, would love to hear more specifics about your findings...
First of all, anybody coming against this Whiteboard Friday simply because it's an opinion that's not necessarily "typical" is missing the greater point. This is healthy discussion whether you agree or not and let's be honest, Penguin is a complete mystery in many ways. Heck, Google (and search engine algorithms in general) are a mystery! Not sure why this is getting downvotes.
To speak to the claim of "link loss," I've almost never seen a healthy, attractive, and active website NOT lose links on a near-daily basis. For many of the reasons that have already been mentioned in the comments, but to me, a website losing links/gaining links "naturally" is a common occurrence that would actually be in an experiment's control, and not be treated as an outlier.
As far as the Disavow tool is concerned, as an SEO, I would never advise a client go about a Penguin recovery without using it, that's for sure. Is it the ONLY way to recover from a hit? Given how complicated website backlink profiles usually are and how large they tend to be, I would say yes. Manual removal is somewhat a waste of time, those link farms and super spammy websites weren't made to field calls and emails from webmasters or "customers," they were made to game an algorithm.
But nonetheless, a Penguin recovery that I've been a part of has always involved a large analysis, some manual removal (when obvious/possible), and then a well-put-together disavow file that covers just about everything else.
Does that mean the disavow is the only thing that recovers you from Penguin? I wouldn't definitively say yes or no. I also think it depends on a website's backlink profile other than the spammy links. Was there enough quality links in the backlink profile already to rank, consistently earn traffic from competitive search queries, etc.? If so, maybe disavow is enough?
But if a small business tried to get ahead of the game by creating a link profile of ONLY bad links, I'd say disavow is not enough. For example, one client we worked with saw strong recovery in rankings for many (not all) industry keywords, but the traffic did not return (all the way) to pre-Penguin levels. That's somewhat due to smaller keyword volumes in their niche, but I also believe it's because they never had a strong link profile, beforehand.
Thanks for bringing this super healthy discussion to a Friday, Josh. I'm not sure there is a definitive answer, either way, to be honest.
my pleasure Brady! I just don't want to see people sold a solution that has no evidence it works.
That's where we'd disagree: "no evidence" definitely seems too strong of a phrase to use, IMO.
Correlation doesn't equal causation, sure, but if (just about) every Penguin recovery heavily involves the disavow tool, let's not ignore that either.
there really is no specific evidence it works. if you have some, submit it :-)
I'm guessing that's because we're not sure we agree with how you're setting up the experiment (even if it is super interesting).
This will be one of the longest and most heated comment chains on the entire site! I have my popcorn ready, glad to see some healthy debate. *For the record...I am in the "don't even use the disavow tool as you can harm more than you help" camp.*
Hey Josh.
Loss of backlinks is something natural in my opinion and it happens all of the time.
For instance: new link to your website from a blog post is visible on the home page of the referring domain when it's new. But when next posts come up the old one goes deeper to older entries. Then Ahrefs or Majestic crawlers while revisiting homepage consider that link as lost. Same thing happens because of the "recent" sidebar widgets and indexed category pages.
You probably won't find the examples of recovered (or even any) websites with literally no link loss at all because of the natural link atrophy.
"You probably won't find the examples of recovered (or even any) websites with literally no link loss at all because of the natural link atrophy."
Why?, As you say a site is winning and loosing links constantly, why this sites loose links before a penguin recovery, and not is like "natural link atrophy"?
The "natural link atrophy" hypothesis is another seo myth which would be great to test! It makes sense, but I can imagine, and over the years seen, plenty of sites that do not lose any links at all (especially in fields where the links are mostly bought). Be that as it may, it is, of course, not my issue if their hypothesis is hard to test! I am not the one making the claim that a pure disavow works ;p I am the one trying to confirm it :-) and I can't. The onus is on them to prove it. However hard it may actually be for them notwithstanding. If they can't prove it, why are they making the claim? PS: their providing easily fakable graphs with a supposed uptick at a supposed penguin release time does not prove an alleged pure disavow conducted prior to this had any effect on that supposed uptick whatsoever. And you notice these people NEVER also provide the link deletion records for that time frame either. PPS: AND I should note I would have gladly accepted the hypothesis confirmed if they only had 100 or so lost links - they didn't. ALL these sites had many hundreds if not thousands of links lost during those time frames.
The "natural link atrophy" hypothesis is not an seo myth. The Chesapeake Digital Preservation Group has published a report on link rot (https://cdm16064.contentdm.oclc.org/ui/custom/default/collection/default/resources/custompages/reportsandpublications/2013LinkRotReport.pdf) that showed that over a five year period (2008-2013) 44% of the URLs they monitored became inaccessible.
More than 90 percent of the top‐level domains in their sample were state‐government (state.[state code].us), organization (.org), and government (.gov) URLs.
Josh:
In order to be a scientific study, you need a control group. "A control group study uses a control group to compare to an experimental group in a test of a causal hypothesis. The control and experimental groups must be identical in all relevant ways except for the introduction of a suspected causal agent into the experimental group. If the suspected causal agent is actually a causal factor of some event, then logic dictates that that event should manifest itself more significantly in the experimental than in the control group. For example, if 'C' causes 'E', when we introduce 'C' into the experimental group but not into the control group, we should find 'E' occurring in the experimental group at a significantly greater rate than in the control group. Significance is measured by relation to chance: if an event is not likely due to chance, then its occurrence is significant". (Source - Skeptics dictionary)
Just because there is a correlation between the 12 recoveries & link loss doesn't mean that disavow only recoveries are a myth. Websites that have triggered the Penguin algorithm have generally gone pretty hard at spammy link building, and have acquired MANY links from the kinds of sites that drop all the time, There is arguably a higher chance of link attrition associated with websites impacted by the Penguin algorithm.
As someone who has performed 300 plus link audits, I can assure you that disavow only recoveries are real. Expecting a website to pop back to its Pre-Penguin position is an unrealistic expectation, as some of the links that once propped the site up no longer have power or may even be toxic.
Insinuating that the whole disavow process is a Google sham and SEO scam is untrue. I know that you like to shake things up, as you did with your blog post "Kings of the Internet | Confessions of a Black Hat" Maybe this is just another opportunity to poke Google & stir the pot.
We posted at the same time Chuck. But we said, essentially the same thing. The sad thing is that many people will watch the video and not read the comments and will assume that because this is on Moz, that it must be true. I don't think these "experiments" proved anything.
hey Chuck, don't tell me that, tell Marie and yourself that, as she is the one making the positive claim All I did was test "pure disavow" results. They all failed that test. 40+ sites now Funny everyone defending the disavow as an option SELLS it as a service... interesting... (and who "me"? stir the pot? naawww ;p ) PS: I don't care if it works one way or the other
Hey Josh - Consider this - some of the people that sell link audits & disavow as a service also have the most experience and arguably the best insight.
BTW - I have also suggested using controversy as a means for attracting links, shares & attention, so carry on :)
ok, where's your proof then? I always said I would submit to proof. So, being the expert you are, surely you have some to provide? (PS: I do link audits and site recoveries too. In fact that's about all I do. Around 300 a year. How many do you do? So we can play the "Listen to me I am an expert" all day long. BUT we shouldn't as the evidence should have the last say IF we can get any - so provide it).
I've been avoiding responding to comments from you Josh, but I felt that I had to comment here.
You do 300 disavows a year? Almost a disavow every single day? Are you doing link removals too? Can you audit an entire link profile in one day? Something is not adding up here!
I can do up to 300 site audits a year, as i said i don't use the disavow file because i can't prove it works, and neither can any of the numerous SEOs who say it does. I've asked them. Including you. i am not referring to you disrespectfully. I tried to warn you that the disavow didn't do anything many times earlier. if you can prove a pure disavow works then by all means do so :-)
however let's not debate that which is besides the point.you say the disavow works. fine prove it :-)
Hey Josh - thanks for the clarification, Now I get it. Since your primary source of income is Link Audits, it works to your advantage to upsell clients on link removal in addition to the link audit and disavow.
Very clever - congrats on the WBF commercial.
not link audits, entire site audits. as i say in the video, you say a pure disavow process works? prove it. post the proof here. Why do you refuse?
I clearly need to up my game. There's no way I can do an entire site audit in one day.
Hi Josh: I'm not attacking you - just pointing out the fact that you have not made your case and that you have not performed a scientific analysis.
And you're right - when i perform a link audit and only a disavow file is required, that's all that i charge a client for. No upsell on unnecessary link removals.
The burden is on you to PROVE that it doesn't work. And stop with the "Prove it - post a url here" red herring. You should know that client confidentiality prohibits that.
BTW - Link audits account for less than 20% of my work, so it's not a life & death matter for me to justify their effectiveness. I'll be just fine if i never do another one.
the burden of proof is not on me. You are the one making the claim that a pure disavow does/might work so prove it PS: fine, email me the urls then - joshbachynski[at]gmail.com
Josh and I don't see eye to eye on a lot of SEO topics; however, I have to agree with him that the burden of Proof is not on Josh but on anyone claiming/selling Disavow as a service. If you're selling someone something, you should be able to prove that it works. It's crazy that no one has actually taken a time to build a test site and put this thing to rest.
Why worry about the client confidentiality Chuck? im sure Josh would be happy to keep the details private. Im interested in why you think Google would spend time and resources to help anyone?
What difference does it make to them if a site gets sunk? It doesnt affect them in the slightest.
Hi Joana: The larger point is the correlation that Josh claims to be scientific evidence is meaningless. The secondary point is that I take client confidentiality seriously. Finally, I'm not sure why you are so trusting of the Penguin Whisperer, profiled here https://sugarrae.com/rants-in-bitchland/penguin-whisperer/
LOL Chuck!
Edit: Moz's comment system put this comment in an odd spot. My LOL was in regards to Chuck congratulating Josh on a good WBF commercial.
I also wanted to make it clear that I am not the spokesperson for the "JUST DO A DISAVOW" campaign. My points here were to say that I don't believe that we have enough evidence either way to say what is needed.
My philosophy is to do everything I can to help a site recover. Disavowing plays a huge role in this, IMO. But there is obviously so much more to the picture.
Marie, Chuck and Josh... I understand you live all this topic very passionately.
But, please, remember TAGFEE.
Sincerely, all this long thread is soon becoming boring, because it seems one of those endeless discussion that, sincerely, I'd like to take place privately.
Thank you .-)
Hi Marie - spot on. This is pure speculation based on a single correlation. I'm guessing that since 90% of the people who die in traffic accidents, brushed their teeth the morning of their demise that Josh would conclude that brushing causes accidents.
Editor note: comment edited for personal attacks
this is getting very transparent PS: read my note below about the scientific process requiring "excluding correlations" vis-a-vi, we can exclude tooth brushing as being causal to car accidents if we do a control test I understand very well how the scientific process works. There is no need to attack me :-)
As Gianluca mentioned, please keep discussions TAGFEE here. Also, please be aware that there are only three levels of replies in the blog comments, so after that third level, it gets difficult to see exactly which comment is replying to which comment, and misunderstanding are much more likely to occur.
Wanted to follow up and let everyone know that I am editing comments with personal attacks to keep things TAGFEE. If you have any questions, please email our team at Moz -- [email protected]. Thanks
Disclaimer: link cleaning, hence disavow file production is not my main service as an SEO, so I am writing this without any polemic intention (i.e.: trying to defend "my business")
I think we are forgetting one thing in this post, one thing that is important:
Google never told that simply using the disavow tool a site can recover from Penguin or a manual penalization.
Sincerely, I never read Matt Cutts first and then others Google's spokepersons (i.e.: John Mu) saying that. I heard the contrary, instead:
Use the disavow tool - and better if you disavow entire domains instead of single URLs - but ALSO show us that you proactively tried your best for deleting, cleaning, nofollowing any backling harming your web site.
Common sense tells me that that phrase is quite easy to understand: disavow files by themselves don't assure a site recover.
But, then, there are also two other reasons that should make reflect any SEO, who really considers himself a real SEO:
First one, what already told by Dan Shure: Google takes a long time for processing the disavow files.
Why? Not because Google is a freaky bastard hating SEOs, but because of how Google works:
We know that googlebot:
Second important point: how the link graph itself works.
Let's say a website has 100 backlinks and 80 are toxic. Google detects those 80 backlinks and the Penguin algorithm applies the filter and the website loose all its rankings.
Now, even if we are disavowing the toxic backlinks, the "positive" effect of those backlinks is not going to be given back to that website, hence the site is not going to jump back in the lost position just because of the disavow file.
That's why paired with cleaning links proactively and using the diavow tool, any hit website should design a strategy for starting earning natural backlinks, which will let the website help regaining the lost rankings.
Conclusion
I understand the reason why this post was written: too many miracle vendors out in the streets (and, ah!, isn't me the one who wrote the post against the 200 Google's Ranking Factors?)
But I don't understand all this discussion and this need to demonstrate if the disavow file alone can lift a Penguin filter or not... if we are SEO and are convinced to understand at least a little of how Google works, from even before ending the question we should answer: no.
yes. also Cutts said to my face, and has said publicly, that a pure disavow will not work because then "spammers can just buy black hat links and then disavow and get out of it"
But it's possible do get out of manual penalty with disavow only. Did that many times and still works.
As Matt-Antonino stated above the reconsideration request may be a very helpful supplement to get out of a manual penalty, and should probably be employed to achieve results...
yes this is true! you can get out of manual penalty with just a disavow ONLY if they think you will NEVER do "black hat" linking again, or it was negative seo against you
Really? So if I send empry RR or just write "please review", they shouldn't lift penalty? Ok will try this in a hour;)
Besides, what about "showing effort" to Google? I saw real efforts, removing links manually and negative answers. IMO because they don't care about "effort" but clean link profile so "manual link removals" are just waste of time and money. Imo still so I'm working with disavow only.
On top of all this, according to Google the disavow file takes 6+ months to take any effect (John Mueller has said this numerous times). So even if it works to any degree, it's no secret it takes forever. Any for anyone who hasn't heard this - Google's explanation is that they have to crawl the pages / domains in your file (on their own independent crawl schedule) before taking your disavow of that page/domain into account.
FYI if you use the Cache Date tool and you ping the links with a tool such as BulkPing, the cache dates get updated MUCH faster. We've had disavow recoveries begin in around 2-3 weeks using this method.
Nice tip!!
you can also submit the backlink page to the webmaster spam reporting tool - it will get that page spidered in 7 days :-)
Hi Josh,
While I don't doubt that many SEOs think they have seen a Penguin penalty removed from them only submitting a disavow, links would have also dropped off at the same time so it will be almost impossible to prove. Coming from a scientific background myself, this is certainly not enough to suggest that you can disavow your way out of a Penguin penalty.
I'd love to see a fully controlled case study on this to see if it can be done! Perhaps something for IMEC Labs?
"Coming from a scientific background myself, this is certainly not enough to suggest that you can disavow your way out of a Penguin penalty." yup
I want to throw another wrench into the gears. Just Tuesday, Alan Bleiweiss received the following from Maile Ohye at Google: "hi alan, checked with webspam manager and a site adding their own domain to the disavow file shouldn't cause harm (it's ignored)." So apparently Google does ignore your own site if you accidentally add it to your disavow file. This supports my theory (unproven) that Google ignores valuable links that you add to your disavow file - i.e. they don't let you hurt yourself. In addition, I don't believe that Google uses the data in disavow files without filtering it - i.e. your site can't be harmed by millions of people putting it in their disavow files alone. I have personal experience that I'm not permitted to talk about with a site that has been disavowed by millions of sites, yet sees no negative impact from that whatsoever. Both of these theories are unproven, and frankly I don't have the time to test them, but they give context for what I'm about to say with regard to this debate.
Penguin is one of hundreds of algorithms (or parts of the overall algo). While we can (mostly) say that a site has been hit by it, we have no clue of the severity of that hit vs. other issues that may be going on with the site. We also can say (mostly) when we think a site has recovered from Penguin, but we have no proof like we do with a manual penalty. In the vast majority of cases, a site that has been hit by a manual penalty, or by Penguin - or Panda for that matter - has been engaging in tactics that are outside of the guidelines in other areas. It's very difficult to isolate specific factors without a clean test environment.
Do I think it's possible that only submitting a disavow file will get a Penguin action removed? Yes, I think it's possible. But only in a very specific case where all of the links pointing to a site are known spam sites and removal is known to be impossible... again something very difficult to test. Do I think a lot of people succeed in getting penalties removed with only a disavow file? No.
I've done a lot of these penalty removals - mostly manual penalties - and (shocker) I don't usually do manual link outreach. I instead look for patterns of bad behavior from the client. We contact people they've bought links from, we cancel contracts, we remove bookmarking spam, and we take down fake blogs. We disavow known spam domains, and usually that's enough. It is my opinion (again, unproven, but supported by a lot of experience) that Google wants to see you clean up your act. They don't need your reconsideration request and your fake excel spreadsheets that show you contacted x site 3 times in 90 days; they just look at your link profile. If you've shown a real effort to clean up your spam, you're usually ok. If you run afoul of the guidelines again, you're in bigger trouble than a disavow could ever get you out of.
Again, this is all conjecture and theory. Completely unproven. It's my belief that it's not possible to prove anything about how Google handles websites since there's always a variable involved (Google) that you can't control. So I'd rather spend my time on helping sites be their best, not chasing algo updates that will change tomorrow.
Loved the video and maybe, more-so, the comments afterward. Way to stick to your guns, Josh!
I have never seen a disavow-only recovery with no new links at all acquired (something that is almost impossible - new links naturally appear every month for practically every decent site, even if they aren't great ones) nor have I seen a site - disavow or no disavow - where backlinks have not disappeared. Links disappear every month naturally and because of algorithmic and manual actions Google takes.
More than anything, I think the disavow tool is being used by Google to find bad sites that their algorithm has not detected and manually de-index them. THAT causes links to drop and that also causes the links to eventually no longer affect the "Penguined" site (another reason recoveries may take such a long time to occur). So, does disavowing links work? Maybe, but not because of the reason everyone thinks it does.
Like almost everything SEO, we all buy into theories with practically ZERO real proof that any of them are reality. I am very skeptical when it comes to anything Google is putting out there. They are looking out for THEIR best interests - not yours - and the ends justify the means in their book.
We all get to have opinions and I'm just happy to see Josh sticking to his! Keep it up, Josh. It brought out a great discussion!
I am both sticky. And I have guns. Wait that didn't come out right....
"I think the disavow tool is being used by Google to find bad sites that their algorithm has not detected and manually de-index them. THAT causes links to drop and that also causes the links to eventually no longer affect the "Penguined" site . . . So, does disavowing links work? Maybe, but not because of the reason everyone thinks it does."
That's an excellent point, and could be the reason doing nothing other than a disavow worked for me. If the sites that were linking to me were de-indexed, I wonder if that means they're no longer a part of my link profile, and are therefore no longer counting against me. I'm not even sure if we can know if that's how it works or not. But if I'm ignorant to something, please enlighten me!
I manage a site that was brand new, and the only links it had were atrocious ones acquired by an outside SEO company they'd hired. When Penguin hit the first time, the site was completely de-indexed for the term they did all the anchor text link-building for. I was unable to successfully remove any of them. They were all from link farms and fake blogs that existed solely for the purpose of link-building. There was no contact info on any of the sites, and the Whois info was always private.
When the disavow tool came out, I disavowed every single one of those links, and never did any additional link building for this site. Not only is the site back in the index for that keyword; it is currently #12.
I'm not trying to prove or disprove the legitimacy of the disavow tool--unfortunately I didn't keep consistent enough records of how many links were gained and lost naturally--but I do think it's quite interesting. From completely de-indexed to #12 with no active action taken except the disavow.
you can check majestic seo or ahrefs - they kept all the records :-) if you have proof of a non-lik loss only disavow penguin recovery, we'd love to see it :-) Also btw, if you were "deindexed" i mean removed from the index entirely, then it likely wasn't just penguin (or peguin at all) as even the hardest penguin does not deindex sites, only burries it to 200-300th or so
I'd agree with this, de-indexation would be a clue to more than just an unnatural, spammy link profile.
I'll check that out. Good to know.
It happened in the same week Penguin came out. And it didn't de-index the entire site. But it did completely de-index it for that one keyword that had all the anchor text. For every other keyword, it came up fine.
While the disavow may not have been the ONLY thing you did, it's VERY clear it's completely necessary to use to ensure a recovery from Penguin. Thanks for sharing.
see i donno about that. i have plenty of sites that escaped from penguin with JUST deletion
Deletion of links? Well... I'd sure like to know how you were able to scale that deletion process! Ha. I guess I should've rephrased "if you can't delete/remove them" because Matt Cutts has said himself that's preferred method.
I've just found that's largely a waste of time for the large majority of the bad links in a website's backlink profile because contacting a human being responsible for that spammy, crap site made to sell links is just impossible.
Another random thought, the disavow tool (if I'm not mistaken) was created to be a proper substitute for when manual removal was truly unavailable or unreachable.
So, while we probably all agree removal is best-case scenario, and you CAN recover from Penguin via manual removal, then why wouldn't you achieve recovery by using the tool Google created to let webmasters tell them "hey, we shouldn't have built this link, please remove it"... ??
Just a thought.
And just to clear it up, I would never tell a client a disavow is all you need. But I'd also inform them about the disavow tool from the get-go knowing that:
Can we please not refer to Penguin as a penalty, it is not.
Referring to Penguin as a penalty is a bit like gaining entry to a prestigious college on fake grades and then when you are found out you are thrown out.
Although I will say the effect of penguin is similar to a penalty.
Ah, Romeo, oh Romeo.... what's in a name? Yes I know Google likes to brand Penguin as a "demotion" and not a penalty... However, as you say, if you are hit by it, it doesn't smell as sweet :-) I'll call a spade a spade in this case Forgive me for not drinking the Google Kool-Aid on this one ;p
It's not about Semantic Josh, it's about clarity and how things should be called.
Even if a Penguin or Panda hit brings conseguences like a manual penalizations, actually they are not penalization, but filters Google runs before rendering the SERPs (filtering & clustering phase of Search). In other words, their fuction is not different than the SafeSearch filter's one.
A penalization is always only manual, because implies the will of punish manipulative behaviors, which are not detected by the spam filters for whatever reason.
i assure you, Cutts has said to my face, and google fully intends, penguin to be punitive. they intend to penalize - whether they call it that or not doesn't matter
that's why I wrote that the consequences are the same as if it was a penalization de facto. But the distinction is important for trying to understand how it works.
Oh, I agree there entirely!!!! When describing your problems to your seo or on a forum, you would best use the proper terminology! I have made that argument myself in the past
Is a site affected by a ranking adjustment caused by Googles EMD algorithm a penalty?
The answer is no, the site never deserved to be where it was, the same applies with Penguin.
This blog is highly regarded in the SEO industry and I don't think we should be muddying the waters. The SEO industry really does not need it.
I can't agree with you on your assertion that Penguin is an actual penalty.
Josh, how do you come to the conclusion from this test that 'deletion of links or loss of links still apparently has to be required in some way'? Just because all of your test samples lost links doesn't necessarily mean that those links had any actual bearing on the recovery of the sites, so what evidence supports your hypothesis that link removal is necessary to recover from Penguin?
I agree that from your test you have been unable to prove that disavows alone will allow a site to recover from Penguin, but you've also been unable to disprove that theory or, from what you've shown, to prove that link removal is actually a necessary component of the process.
the fact that they all had link loss, in fact a ton of link loss, the more link loss, the higher correlation with recovery ergo, i cannot say a pure disavow works. nor can anyone else. as no pure disavow happened. all i can say is a disavow with major link loss happened, and they seemed to recover... in fact some had quite the link loss... ergo major link loss might be a factor in recovering from penguin PS: please do not say to me "well that's all correlation" - Of course it is. ALL we have is strong or weak correlation tested in ways to exclude other factors! That's scientific testing. there is no such thing as causation* - there is only levels of correlation. it is that people make unscientific conclusions on weak, non-excluding, forms of correlation is the issue *as far as we can tell (the only way to tell it would be correlation). this has been the epistemic problem of science since the early Greeks
"the fact that they all had link loss, in fact a ton of link loss, the more link loss, the higher correlation with recovery"
Well that does change matters. If the higher the link loss, the more 'complete' the recovery then that would start to suggest some connection.
"ergo major link loss might be a factor in recovering from penguin" is probably how your conclusion should have started to begin with (along with more information on the size of link losses involved to demonstrate that it's a significant and consistent trend amongst the sites sampled).
And I don't think anyone would argue with that; removal of bad links has been a given if you want to recover from link penalties (or Penguin) since long before the disavow tool existed. I would think that it's still possible to recover from link penalties and Penguin just by removing links, provided you were able to remove enough of the bad links.
The question which still remains is whether disavowing links is merely a placebo, as you claim, or whether it has the same - or similar - effect as removing links in aiding recovery from Penguin.
You haven't given any reasoning or evidence to justify your claims that using disavow 'is a placebo' or 'at BEST a dubious practice, and at worst a hurtful one'.
Just because there's some evidence in the situations you've examined suggesting that the disavow might not have been the lone reason these sites recovered doesn't mean that is the same for every case, nor that the disavows had no effect at all (or damaged the sites).
If you're going to take the consistent link loss as evidence that this might aid in Penguin recovery, how can you at the same time reject disavow requests as something that might aid in Penguin recovery? Both of them were consistent factors across all 12 sample sites.
Should the disavow not be used at all? Should it be used in conjunction with link removal? Is link removal alone sufficient to recover from Penguin, even if you can't remove all/some/most of the bad links?
All that we can really learn from this is that from the 12 sites selected, link loss or disavow requests or a combination of both might have led to them recovering from Penguin.
You certainly haven't disproved that pure disavows won't recover a site from Penguin - in order to do that you'd need to have a significant number of sites which experienced no link loss, did disavow all their spam links and didn't recover from Penguin.
So your conclusion at this point should really be "major link loss and/or disavowing links might be a factor in recovering from penguin".
Well stated, Ben.
In other words, there was absolutely no conclusion to this "experiment". In the future, I think that Moz should limit WBF's to presentations that have some useful information, not merely "I was unable to conclude anything". What are we supposed to do with that?
Many things. Don't rely on the disavow. Don't sell it to clients saying you know it works when you don't know that for sure. Don't make or perpetuate the cause correlation fallacy. Don't assume SEO is based on hear-say and myth. Don't buy the Google hype. Etc. Not to say you are doing any of these things. But why do you assume every WBF has to give you easy tips? Maybe some SEO issues are not so easy ;p but still need to be discussed? I don't understand this hostility.
Hey Josh,
Let me explain:
First of all, I certainly did not mean anything personal.
I've enjoyed your presentations in the past and I hope you will continue to share your knowledge and insights with the Moz community.
I was just put off by your combative attitude in the comments here and in the WBF.
The Moz community is a group of intelligent, educated, and honest people. Google makes it very clear that the disavow tool is not a magic wand that is guaranteed to yield results. Like much of the information that SEO is based on, it's a Google recommendation that can neither be proven nor disproven. So any intelligent SEO will at least play it safe and include disavow in their arsenal of tools.
This is what we knew before your WBF. All you added to the discussion was wild accusations (that Google is scamming us and that SEOs are scamming their customers) that have no basis, and unscientific assumptions ("So I'm going to say that I think the myth is busted." You didn't disprove anything). Charging money for a service that Google claims has value and SEO's feel works (yes, I believe this is very significant! The people who are doing this actually believe in the service and find it to be useful.) is not the same as maliciously scamming people out of their money, and your suggestion that it is comes across as slightly arrogant and insulting.
To be clear, I am an in-house SEO. I have nothing to gain or lose from this discussion. I was just disturbed because I feel that this kind of rhetoric is not on par with the level of the Moz community.
Respectfully,
Yair
I am very happy if the people selling a disavow with no evidence are insulted - they should be. They are RIPPING OFF clients! I am defending people here... why am I the bad guy? Notice above I said "if" and "no evidence". If they have evidence I challenge them politely to proved it I have done nothing wrong here and my conscience is clear, despite any feather that may have gotten ruffled :-)
Your test would be more useful and credible if you included the number of links removed, even if just a broad relative-frequency would be better than nothing.
Regardless of the outcome I'm firmly in the camp of removals, we don't know how things will be handled in the future so just remove it to be safe.
Your comment on Google's use for this data; I don't think we need Googles confirmation for machine/deep learning, it's obvious - it's rich snippets all over again.
In doing so I would have to submit their URLs... can't do that :-) and make for a very dry WBF
Not the domains just the figures
1) I never saved the numbers as the number weren't important. It was a binary question: "Was there any considerable link loss?" (I would have ignored anything that had like only 5 or 10 links lost or some small amount). In all cases the answer was yes: major links were lost 2) Either you believe me or don't. I can fake screenshots just as easily as I can lie and say yes major links were lost. *** We have a disease in this industry where screenshots (which can be easily faked) are taken as gospel truth *** This is why I didn't provide any here. They are not required. They are actually hurtful. Repeated Peer Confirmation would be the true way to prove it anyways.
It's not a question of belief... It's about understanding the experiment.
You talk about losses, is it 5 links, 500 links, 5,000 links? You've now vaguely answered the question now.
Just an FYI please don't take this as attacking you, just trying to understand your thought process.
Even if the experiment showed disavowing was all that mattered I'd still manually review all links.
i understand :-) np - still in my mind it was a yes or no thing - they say absolutely no link deletion efforts are required - i wanted to confirm that hypothesis if i could. i have not been able to to date. neither can they. so until that time, i say we can't say, it works
Good point on this whiteboard. I am also trying to figure out if the disavow itself really works in any way or not. But I don't think so. I support the assertion of Josh that the myth is busted.
To say the disavow tool "doesn't work (at all)" would be ignoring LOTS of evidence from LOTS of experienced SEOs. It may not be the only thing necessary, but it may also be necessary for recovery. Unless you can get all those spammy sites to respond / manually remove awful links. Good luck with that, btw! I've tried!
i'm sorry brady -- and you might just want to be diplomatic here :-) -- but there is no evidence yet submitted to the me or the general public that a pure disavow works :-) i've tested it all. i can only go on the evidence i have access to, however "experienced" other SEOs claim to be. and yes I have seen sites with only link deletion and no disavow at all recover from penguin. yes it is virtually impossible to do effective link cleanup when it is an uncontrolled set of links when it is a controlled set of (black hat) links, it is very easy to do ;p so that's why i have seen it a lot Another interesting point... and that is also why other SEOs sell the disavow, b/c otherwise what are they going to sell to their afraid clients? link deletion doesn't work so well, and is costly, so they don't even try and instead just sell them a dubious disavow service that they don't know for sure will work or at least, that would be a plausible hypothesis ;p I don't really blame them either as long as they tell the client it is a very long shot. maybe changing domains, or starting another better site, would be better
Yeah, I would never say that you can't recover by manually removing links. That's best-case scenario straight from the horse's mouth: we all know that.
But just because a "pure disavow" isn't showing you the evidence within this experiment doesn't mean it "doesn't work" at all. That'd be a silly claim, IMO.
Is it the only thing you need to do? Probably not, but it still AIDS recovery and thus, works.
Using the disavow tool helps sites recover from Google Penguin. As a substitute to manual link removal, I can't see how it doesn't help sites recover. But that's just me.
If an SEO is selling "just disavow and done!" then they're wrong. Plain wrong. They cannot guarantee that nor can they know anything without seeing a specific website, its backlink profile, etc.
I just can't get behind your theory that the disavow tool flat out "doesn't work," even if it does make the discussion more entertaining.
:) Thanks again, Josh.
Editor note: comment edited for personal attacks
I'm nothing if not entertaining :-)
It works, I'm working only with disavow and don't care about manual link removal. IMO wasted money.
Hey Josh,
Just clarified this with John Mueller. It seems that, in the case of Penguin, simply disavowing the "bad" links would be sufficient to remove the penalty. Any other steps taken in the direction of cleaning these links might help, but aren't mandatory.
Also, physically removing the links is technically equivalent to disavowing them.
Source: https://youtu.be/6p1igrDVqC0?t=43m20s
Thank you Mihai for posting this. I was beginning to feel the tug of TLDR when I found your updated post. It would be nice if John came in here to comment...does he frequent this community?
Google doesn't have tools to "check You" if You really did link removal requests etc... It's not possible and it's not possible to not get out of penalty with disavow only. I don't have control of some links harm my website (doesn't matter made by me or not - Google can't check that) so I use disavow which is working fine.
Also if disavow isn't enough, manual penalties shouldn't be lifted just with disavow only (I'm working with this only and for now over 100 manuals were lifted). Same thing is with penguin but it takes time (or not - orca, got results too and not just one). This will be possible (disavow or that wasting time manual requests which I think it's just to earn some extra cash by "specialists") only if number of "bad" links isn't too big. I saw many clients with 70-90% pure money keyword, tons of unnatural links etc to disavow. It worked but they (almost 100%) think "lifting penguin is the same as getting back to top10". This isn't possible because number of "good" links left is low. However overall serps/traffic/impressions rising over time but not to previous "top10s". Is it penalty lift? In their opinion no but in mine yes. No top10s but google allows website appear highier without changing content, earning new links or something similar. And this can be made by disavow only.
Disavow works (manual requests too but this is imo to get more cash) but don't expect top10 if You have 80% bad links;)
I think you have an excellent point in mentioning that people need to consider the fact that their pre-penguin rank may only be attainable, after performing a disavow or removing "bad" links, when adding new quality links.
It's a must. 1000 links, 700 bad and 300 aren't strong enough. Overall serps/traffic/impressions rised but not to top10. They say penalty isn't lifted but I say yes. Why? For example traffic rised 60% and impressions 200% without adding links, changing content etc. Some clients can't understand this or don't want even if You have strong and logical arguments. They say: top10 and penalty is lifted!:)
@Josh
Thank you for the post, hope many read this. I hear all the time about all these theories and here says that can get you out of the Panda nest. So much of it can be misleading and harm your efforts even more.
One thing many miss is that people want facts, not theory, unless they are facts supported by theory.
Awesome post thank you
Josh,
While I don't think that you are wrong when it comes to the disavow tool I think the way that you are testing your hypothesis is incorrect.
As someone who works with a lot of data I think that drawing your conclusions from 12 tested sites that don't fit your criteria is not the best way to reach your results. Some more controlled testing would probably be needed before you could draw any conclusions.
Have you asked for any examples of any sites that have not lost any links, disavowed, and not received recovered? I haven't seen any at all. I feel like you're taking a limited view at data, and you are coming to conclusions without holding enough variables fixed.
It's hard enough for me to find a whole site that isn't losing links. The other problem is most spammy links tend to drop off a lot, for instance spammy blog comments often are removed, or the sites they are placed on will often go under.
I don't disagree with the idea that the disavow tool really isn't an easy fix like most people think. What I disagree with is the way in which you reached your conclusion. Your "scientific" testing is flawed from what I see.
Hey Nils-Evensen! Thanks for the comment. However I have to disagree a bit :-) You see, I am not the one making the positive claim that a pure disavow works. Other people are. I am asking, where is their proof? Until they provide some that a PURE disavow works, we cannot say the pure disavow works. Nor should they. It is not my problem if their hypothesis is hard to prove. That's their problem, as of course, the burden of proof is on them. You see? Besides, it is also speculation on our part if we say it is hard to prove because "all sites have link loss". I know plenty of sites that don't have any link loss at all.
Man, have I had fun reading these comments. And I'm noticing a theme. Josh keeps saying, "I'm not the one making these claims, they are. So it's on them to prove it." That makes sense on the surface. Except Josh is the one doing the experiment here. When you do an experiment, the whole idea is to form an hypothesis then test it.
As much as anyone claiming that the pure-disavow does work needs to provide proof, as do you that is does not. And their inability to provide proof in no way proves your theory correct. And vice versa.
I just don't understand the methodology in gathering your data. You decided to conduct an experiment and relied on asking people to send you "evidence" to disprove your theory. When no one could, you proclaimed your theory correct. Or at least the opposing theory "not confirmed." That just leaves me unsatisfied with your proof.
Is data from 12 websites enough to confirm your hypothesis? Are you going about the experiment in the right way to prove your hypothesis? I think these are the questions Nils-Evensen is getting at.
You say that you come across sites all the time that almost never lose links. So why aren't those sites in the experiment? I assume it's because sites that are never losing links and always gaining them are a lot less likely to be hit by Penguin because they have a healthy link profile (are making content people want to link to permanently).
I'm all for the righteous battle against shady SEOs. So, I'm fine with trying to shame the ones that try to sell the Disavow tool as the one-and-only way to fix link problems. But, are you saying that there is NEVER any reason to ever use the Disavow tool? Or just that ONLY using the Disavow tool won't get you off of penalty?
Because unless a lot of commenters here are just lying, it seem like a lot of people have used only the Disavow tool and recovered. So, just like you have recovered sites using only manual link removal, they have done it using only the disavow tool. I know you are saying that all the evidence you have looked at shows a loss of links associated with the recovery of these disavow-only methods. But, if we take them at their word, and no manual link removal was done or attempted, what is going on there? Are they all lying?
As far as recovery methods go, I always advise people to use every method mentioned by Google to try to appease them enough to get off of the penalty (or recover from the effects of an algo update). And Google has made it pretty clear and simple - try to manually remove all spammy links and those you cannot get removed, add them to your disavow file. To me, proving that it works is less important than the fact that Google advises you to do it. So, in that regard, I can't see how any SEO involved in penalty recoveries and link clean-up can tell a client not to use the Disavow tool. When the client say, "But what about all these quotes I found from Google saying you should use it when you can't get links manually removed?" I, for one, will likely not respond with, "Because I can't prove that it works." And, at the end of the day, it certainly won't hurt a site to disavow all of it's crappy links.
Well, well, well... Finally someone got the balls here to say it out loud: the disavow tool doesn't f work!
Ain't no worries about so many thumbs down, Josh- you're only trying to challenge their livelihood.
Thanks for the article and keep up doing great work, bro.
I'd say you're right- that disavow isn't the be all, end all solution. It would be nice if it was- but that's too easy isn't it? We know nothing is that easy when it comes to Google.
Hello Josh... Interesting article. Forgive me if I'm wrong. What If Disavow actually means "losing links" in Google's perspective. I guess the Penguin recovery is confirmed solely using Google SERP performance right? If we ignore the third party software's link analysis which I believe is no way near the way Google sees & grades your website, and for a second pretend they don't exist, and then see what happened... Its more like "I disavowed suspicious links, and It worked" !!!! I think the best way to see your website like Google does, is by using GWMT. It might not be accurate as other SEO tools, but it doesn't matter, life ain't fair right? What's your take on using "Links to your site" option on GWMT for the same experiment you conducted? Your thoughts Josh? Thanks in Advance.
Hello Everyone,
Just to add here after watching this video and going through all these comments my opinion relates to Marie and a lot of fellows over here who are in favor of Disavow tool. But, I agree with you as well Josh!
There is a simple thing as I've learnt from a number of tests we've done for a number of websites that recovery from Penguin (filtered or manual) needs following
- Thorough research on links that are pointing to a website (either built, gained/purchased, link bait or competitor's attack with negative links for you, or any other)
- Understanding the nature of links (Good or Bad), as Rand said in one of his very earlier White Board Friday's video after the penguin was launched for the first time.
- Filtering out all those bad links
- Link Pruning (If you have a control on it or by requesting removals)
- Preparing a Disavow sheet for the links that you can't delete or don't have control of it. Instead of leaving it as "can't be removed" for the reason as YOU say Disavow doesn't works, why not we add it to the Disavow sheet ??? Even if someone has not proved it scientifically, we still have "HOPE" that it works.
So in my opinion, both exercises "manual removal" and "disavow" are needed to hope for a win win situation.
Spur of debate and a needed and good one. I cant fully agree to this WBF but i cant ignore it as well. While there is no evidence of what Josh is claiming I fail to see anyone bringing a site as an example of what opposes it. But purely writing that disavow alone fixes your problem is not likely and we should all leave it at that rather prove different points.
late to the party here as I've been away but IMO this is a near impossible hypothesis to prove as lots of affected sites will have a pretty high natural attrition of links anyway, as a large number of directories articles sites whatever other rubbish is out there are simply not being renewed or maintained as they can’t make money they used to post panda/penguin. Therefore sites will be losing crap links on a almost daily basis anyway.
I have seen recovery both ways but impossible to 100% say it was just disavow when links are dropping off constantly, particularly when there is 12 months between refreshes, a hell of a lot of links can drop off in that time.
How I pray for the day people in this industry stop trying to find tricks, fool google, and do clever things with code to avoid real work.
This entire discussion does little but prove the prevailing mindset is still to take exactly the same path that got people into trouble in the first place.
It makes me sad :(
Are there really a lot of SEOs claiming that you can get out of a penguin problem with just the disavow tool and no links removed or built?
Maybe I have not been looking at enough shady SEO sites to see that, but this seems like a straw man argument.
that is a valid point - but read the comments and you will see :-) also read their posts and numerous tweets. these guys aren't "shady" or they pretend not to be. They are white hat. Some of them even post frequently on this forum! Hence why so many people disliked what I was saying here lol ;p !!!!!
Hm. Just not seeing that. I see some anecdotes about a disavow-only recovery, but even the people who really took issue with your assertion don't claim a disavow-only approach is a good idea. For the most part, it looks like people disliked your methodology and the assumptions that came from it.
It is very hard to prove something does NOT exist or work.
Hate to take it off topic, but this kind of reminds me of how people attacked me for saying that negative SEO is not common, not easy, and isn't something most people need to worry about. People who had something to gain from fear mongering about it really had a problem with that. Just like you have been doing here, I asked the true believers for proof in the form of examples which didn't have a lot of their own link spam, and people just pointed me to hype-filled fact-deficient articles about how it might work.
Here's where that story gets more relevant to the Disavow topic:
I have had over 2 years worth of linkspam and other nastiness aimed at my site, some of it from people who just didn't agree. It didn't really have any noticeable effect at any point in that timeline. After discovering what was happening, I disavowed the bad links, sometimes before GWT found them. Didn't bother doing removal requests. After a few months, I said screw it and stopped disavowing, though most of the worst links had already been disavowed by that time. Eventually I did add the remaining bad links to the disavow list just before the most recent Penguin update, just in case.
If you want to use that isolated case as "proof" of something, it would mean that either the disavow tool works and prevented a penguin penalty, or negative SEO does not work.
all of the non-relevant controversy aside, it is very simple, people have said to me (and they have said it, whether they wish to admit it now) that a pure disavow does work and tweeted graphs to prove it. I have asked them for the urls. 40+ now have responded. I checked and there was link loss. Therefore, there claim of a pure disavow is false. Until such time as evidence can be submitted it cannot be substantiated.
Who said a pure disavow works? Name some names. Prove it! ;P
I'm in enough trouble without naming names :-) but there are some people who have in this community in fact they have even responded in this comment thread
Great post Josh...after going through 137 comments and a very interesting post it appears that Josh question still remain unanswered. If someone had ever lifted penguin penalty only by using Disavow tool then please share the URL. I want to share that I had used disavow tool along with possible link removal campaign for two of my clients at the same time and had seen different results. I managed to lift the penalty for one client( it website was a .edu domain & a well known institution operating from last 20 years) a bit early than other. From my little experience i learned that disavow tool works in different manner for different niches. Josh I am mailing you the URL's just have a look if possible.
Read the blog and checked the video as well. Although I am quite late but just wanted to say few things here. First thing to the Moz community.
Dear Moz I have been your follower since the time you were SEOMOZ and still I follow you. But what's wrong with you? Why are you publishing articles which are misleading and contradictory when you have a good R&D team sitting at your office. I mean Rand what's going on, why you people are continuously degrading your research and content. Try to understand you are a brand and people like me still look forward to learn from you and your website and the way this article has been published I am very disappointed.
Now the article. I would like to put it in the strongest possible words, Yes Only disavow can help you to recover from penguin. But before any of you say anything read this.
Make sure you understand what is Disavow? Check the Google guidelines and then fire on me. Disavow means first you work hard on removing the links, once done then you submit a reconsider request and believe me it works every time. Not sure how many of you tried this but I being one of the senior authorities at my company have seen this for years. In fact I have run more experiments on this than many of you here and it works fine every single time if you abide by the Google guidelines and ready to work hard.
Just wanted to say I am not here to offend any one but I always feel bad when things go bad and something is misinterpreted in the industry and that too at a big platform like Moz.
Ankit,
Thanks for expressing so well what so many of us are feeling about this post.
Hi Josh, great discussion on disavow process and its effect when implemented as recommended by Google. We checked out this process on 3 clients but still trying to figure out if the disavow itself really works in any way or had any impact... Clearly when we checked Majestic SEO, we can see the link loss still and stuff...
After several attempts using the disavow process, my team and I came to the conclusion that the disavow process is just another MYTH that seo's have to deal with. Thanks to the Penguin for giving us more work and leading us to a complete dead end with the disavow process
Josh, thanks for shedding light on this and starting the process of busting this myth for GOOD.
Alex
Hallo
Hey Josh...The domains which you tested had manual penalty or algorithm penalty?
I'm happy that you made the important point at the end which is that companies should stop working with trendy SEOs and should work with those that focus on data driven decisions.
However, I see a hole in your analysis. Majestic.com's link loss report is a poor data source for identifying link loss for 2 reasons: 1) it never seems to report the metrics correctly due to their small index and for reasons unknown, and 2) links loss is a common occurrence especially if the link was from a blog comment or article that was circulating across all pages on a website.
You may have figured out a way to isolate these issues. But they are worth noting.
I understand that you cannot factually identify a disavow file working, but looking at lost links is not a good way to take your approach either.
We need to run another test. Right now i have a client who needs some links removed/disavowed. I will run a ahrefs and a majestic + moz OSE report today and submit my disavow file today. I will then need to wait and see if this helps with rankings since no matter what we do he will not move back up to page 1. Email me so i can send you the data and get this on paper.
I can tell you that its been 9 months of trying to remove links and add quality links. He has barely moved. He does have way too many identical anchor text links and that is a penguin also penalty IMO. These sites are unresponsive to our requests. The disavow file is having a list of all these directories and sites which are linking with similar anchor text keywords. The rest of the links were untouched since they are random anchor or simple branded + www links. We have tons of citations as well.
Any other questions i should answer before we can use this test as an example of disavow file working for bad linkage/anchors?
I have also used Disavow tool and the result was so instant that my site again ranked on 1st page with better ranking than before Penguin effect. Is it happened by Disavow tool or naturally?
I have one site that recovered from Penguin last Oct, and I didn't disawow anything. I just obtained some extra new quality links
Great post Josh. Even though we always recommend using the disavow tool I appreciate your skepticism and not allows SEOs to just fall in line with whatever Google tells us to do.
I filtered all backlinking domains for Page Rank, then those of the lowest PR for PA/DA with bulk checking tools. This should narrow them to those of the lowest quality. This is best i could came with for most efficient way of weeding mass bad links, after reading all there is about disavow.
Hi Josh,
Thanks for the video and controversial discussion. I put up my 2 cents regarding this as well, and a penalized site's stats that have recovered.
https://chiujinghong.com/does-google-disavow-really-work/
Question:For some people talking in hints: Does Disavowing links works? Is Penguin recovery possible?
Answer:
I've recovered site before. The big problem people face is getting an accurate list of backlinks. Webmaster tools is simply not enough. You have to use all sources, ahrefs, open site etc. I saw signs of recovery after 10 to 20 days.
You have to keep up the search for backlinks. Like on a weekly basis.
It is possible tho?
You should also take the opportunity to look at the site layout and structure. Im noticing that its very easy to get a punch from panda in the last 3/4 months. And if you have been hit from Penguin i reckon this flags your site for a panda evaluation. The two updates imo are more linked and intertwined them people realize.
My opinion is that it is impossible to tell, because the penguin recovery process takes time and in that time low quality links will be dropped, so any site that is affected by penguin will have likely had many low quality links pointing to it dropped. So, you will never find a site affected by penguin which doesn't have dropped links that were pointing to it. so you will never know if the disavow process worked on its own or not. So nobody will ever be able to show you the poof you are looking for.
The company I work for got hit by a competitor on Christmas eve. Our top keywords went from (1-3) -- to 60+ in less than a week, with no manual penalty. I added over root 300 domains to my disavow. 12,000+ new links all pointing to the same page, with an odd twist:
They spelled the company name wrong (essentially changed the niche).
It would be impossible to email all of these domains/sub-domains webmasters, so I am relying mostly on the disavow. The company's' fate depends on it.
Side note: I also thought Google claimed it would be very difficult for a 3rd party to negatively affect SEO. Definitely not the case after an extensive review.
Hi CBuettner,
The most important thing in your situation is to make absolutely sure that there are no manual actions applied - you can do that by checking the Manual Actions Checker (under Search Traffic in the sidebar).
Once you know that there are no manual actions in play you can more confidently manage the situation with ongoing monitoring of incoming links and regularly updated disavow files. This information about using Disavow files in my deck from SMX East might be useful for you if you are going to need to employ this as an ongoing strategy. You can find it here
If there is a manual action in place, obviously the approach needs to be different. You will find some help in that same deck for dealing with those as well.
It would also be a good idea to make Google aware of what is happening with your site - the more they can see actual examples, the more chance that they may accept and adjust for Negative SEO affects. Here is a link to the Webmaster Central Help Forum
Hope that helps,
Sha.
Hi Josh !
I agree with your article, but Penguin still holds a lot of gray areas.
Great post :)
Let's not forget that Penguin 3.0 refresh was specifically aimed at sites who have had their links cleaned up since the last refresh and any links that couldn't be cleaned up or "died" of natural causes were added to the disavow file. You may have just added a disavow file a month before the refresh for the first time in that year between refreshes and found that your site recovered but that was only down to a combination of lost links and Penguin taking into account the spammy "links" found in the disavow file.
Let's be clear too, you would not have recovered if you had or still have a considerable number of low value links that you had not added to the disavow file a month or so before the refresh.
There are web owners out there who have seen there sites drop because they have a large number of spammy links, which they may have created themselves during the last few years when it was easy to submit links to directories and who know nothing about the disavow file. They probably rely heavily on online business and these are the businesses Google are targeting to switch to Adwords campaigns.
Without an SEO community providing legitimate Penguin recovery services for these types of businesses then they will forever be in the grasp of GOOGLE.
We've a few clients with not just a link penalty but there last agencies all just did really dumb link building...anyway to cut a long story short a couple of months back the rankings just returned which I was pretty shocked about, in the last year and half I've spent plenty time running around sorting out penguin related issue all to not much avail and suddenly the penalties are just gone... the only thing I can trace it back to is a change of mentality at Google to disavow requests
i would say that we cannot rely only on disavow tool, we need to remove those bad links. and as per my experience if you have number of spammy links then you may need to do few reconsideration request. but one thing is for sure that the result after the recovery will be amazing. so keep working on bad link removal process.
This is great food for thought. I think its time to look through the clients who we just recovered from Pengiun and see if more happened than just the Disavow...
Hi Josh,
I might be missing the point of the post but I wanted to comment as the other comments were so good. I don't that anyone is disagreeing with the outcome of the research that you completed but more the way you came to the conclusion. It has always been said by the powers that be at Google that a disavow should be used in conjunction with manually removing links.
It was a very small dataset to work on and I think that you are always going to get fallout of links over time which would cloud the results, but I am just going over old comments.
Though not having a penguin penalty for any of my 35+ clients going in to 3.0, I did do a disavow for each client in preparation. It wasn't very heavy handed but instead getting rid of some old links that are now considered no goes. Only one of my clients saw a hit from penguin in October.
What is the bast way or tool to check your bad backlinks?
The absolute best way (though not really practical) is to get a list of all the links you can find and then view each link individually to see what it is, where it came from and whether it's good, bad or fairly neutral.
That being said, that's not normally how it's done. You will need a list of all your links. To get those, download your webmasters tools links, use ahrefs.com and majestic.com along with Open Site Explorer. Those 4 sources will give you the best chance at finding all your inbound links.
Once that's done, you need a way to sort them - you can either sort by some Ahrefs or Majestic stat or you can do your own custom work in a program like Excel (ie. pulling in Pagerank, Page Authority, etc.)
There's no tool which will do 100% of work. There're tools gathering links and data based on ip, c class, anchor, outgoing links, ga code, etc. Then You need to select what is good and bad. Without experience You could select too much good links or not enough bad. And mess up Your serps.
Use SEO Spyglass software paid version that is an easy and affordable way.
Easy, but at the end man should be to double check (experience is required...).
I'm a huge fan of this method: https://www.greenlaneseo.com/blog/2014/01/step-by-step-disavow-process/
Gather data from ose? Majestic and ahrefs got more every time I'm checking, so I'm not recommending ose as a link source. Cognitive is a tool, just another handy tool but won't do all for You. Experience is needed, still.
Oh and "making effort" isn't needed if link profile is clean.
I believe that disavow works for one reason and one reason only: Hope!
Hope that negative SEO cannot affect us, or at least cannot destroy us for good!
Because if disavow is a myth, when you terminate your cooperation with your SEO agency, they can simply attack your rankings with bad link-building techniques in order to rehire them!
With this in mind, that is a huge thread and you can tell it from the comments as well.
Disavow works fine but many people selling "link removals" won't tell You that because they could loose some extra cash. Or just will tell You "better remove links manually". No matter how much links they'll remove, You have to pay it. So why pay if disavow works?
It makes sense, from the perspective of Google, that disavowing would remove a penalty but NOT recover your rankings. That is, you won't have that algorithm holding you back anymore, but you have to rebuild trust.
Should we still be talking about penguin as a penalty? Penguin is an algorithm unlike a manual links penalty that is handed out by a reviewer. Now that it has shifted to continuous updates it’s just part of the search mix and another signal that SEOs need to fix in order to gain favour with Google. Fact is if the SEOs of site owners have realised that whatever activity they have done in the past could be harming the site, they’ll have stopped said activity so link numbers are bound to drop. Loosing link equity through natural link loss and disavowing at the same time, you won’t see a ‘recovery’ so to speak as the links were probably not the types of links that should have been holding up the profile in the first place.
As SEOs working for clients we only have a finite budget to work with, in this case could you afford NOT to disavow what we know is toxic to help protect the site as best we can? My thinking is that the disavow is used to tag up a site to help tell Google that is shouldn’t be used to count negatively against the profile in future, yes we may not know that explicitly but can you afford NOT to do it?
As long as Google holds the market share of organic search to a site we need to do whatever we can to protect search traffic from this source, even if it means helping them with their algorithm by disavowing.
Link auditing, like technical auditing, content auditing etc. is now just another part of the online mix to ensure that the foundations are laid in order to grow the site.
Yes disavowing alone may not immediately help recover the site but it will hopefully protect it from negative link affects. Funnel the rest of the budget into actually creating a site that engages a customer and meets their needs.
That is ultimately what Google wants. Win that race and you’ll win the rankings.
Hey Alan! thanks for your thoughtful response. I understand your point "how can you afford not to use the disavow" However I have to disagree with that thought just a bit. My issue is that 1) if the disavow does nothing then it is a waste of time, and money, and more importantly hope. But, if it works as you say it does, then it will actually hurt your rankings as you will remove those links from your link graph on the next penguin data refresh as Cyrus postulates he might have. Plus, many SEOs arrogantly assume they can select the "bad" links like a surgeon (yes, imagine Weird Al when I say that) and leave ONLY your good links intact. They will inevitably remove some good links from your link graph as well. Perhaps LOTS of good links. Because no one, and I mean NO ONE, knows what is truly "good" or "spammy" in that regard. It is a high risk maneuver only to be done when you have absolutely no choice (like with a manual action vs site due to links - not the partial - which you have to do nothing - if you remove any links then you are just willingly selecting more links to remove from your link graph - you are neg seo'ing yourself!). At any rate, many people have seen ranking drops after penalty or demotion when they used the disavow tool. In that case, if the disavow worked, then the cure is worse than the disease!!! 2) we have SEOs companies and consultants selling the disavow Also we have SEOs whose full time job is now looking for "bad" links and putting that into a text file!!! this again is a waste of precious time, money, effort, and when you should be making your site awesome or doing social or ANYTHING other than filling a disavow file that likely does nothing or just might end up hurting you!!!!!!!! What happens next year when google drops the dofollow link signal almost entirely? (yes this is my prediction) So as far as we can tell people are spending money on this when they could spend that precious time and money on some other activity that may actually help their ranking. Without deletion they will apparently not get out of penguin. So, if and only if you were hit on these google confirmed penguin dates: 2012:apr 24, may 25, oct 5 2013:may 23, oct 4 2014:Oct 17-18, Nov 27, Dec 2, Dec 5, Dec 6 invest in link deletion, or change domain names, or delete the offending page on your site (if possible) or try a tricky way to get out of it (email me), but don't use that "disavow crack" as it doesn't really seem to do anything BUT waste your money and possibly hurt your rankings. Or as a last resort use the disavow file (if it is one time and don't get sucked into doing it daily) It is a placebo that people believe in because it feels good :-) But there is no proof it does anything, and that will end up hurting sites and people
Yeah I agree. From my personal experience just disavowing the links will not help. You have to go through whole process link removal, documenting link removal, disavowing, sending recovery request.
I think you are confusing penguin with a manual penalty. You can't submit any documentation or a reconsideration request to recover from algorithmic actions...
I think we cant disavow out of penguin just that easily, as penguin takes place with big data refreshes so the only thing we got there is to wait for the big refresh to take place. I have personally disavowed 4 to 5 sites but they recovered only after the major refreshes.
what are the urls?
I am planning to write on this topic, will be providing URLs with complete analysis :)
Hi Josh
Wow - I was always under the impression from what I have read that simply disavowing would work.
Just wonder what to do for negative SEO as I think thats the biggest area of concern of 2015 so just wondering what your tips would be to help in combating negative SEO. If your saying disavow alone doesn't work, what would be your suggestions.
Thanks
Andy
link based negative seo: try to get those links taken down if possible through legal means or legwork - if they look really damning otherwise we are in a terrible position where we have no choice but to use the disavow which we cannot confirm works but it in many cases is all we got yes i think this is terrible. and google's fault. and they don't care. there are rumors that google DOES take the disavow into consideration when and if the manual team is called in to look at suspect links... for that we will simply have to take them at their word. PS: There are some other tricky things you can do: 1) make sure to use a canonical on any URL to your canonical version (including parameters and HASH based urls) - this way it will be harder to negative seo you on the panda side of things (yes, you CAN hit someone with negative seo on the PANDA side - email me if that doesn't make any sense... maybe that is an idea for a WBF... ;p ) 2) there are other tricky ways you can use code to get around link based neg seo attacks. I can't release them publicly as I am currently using them to save certain sites from neg seo attacks like I mention in my 2015 SEO course here: youtube.com/jbachyns (time 36:29) PPS: there are a ton of other negative seo tactics that have NOTHING to do with links - see some of them in the same video youtube.com/jbachyns (time 59:14)
I know something I won't tell... riiiight...
What's that? 301s of bad domains/negative seo? I don't know what's the name of it but something with showing other url (cloaking?) after "clicking" negative seo link? If yes, that's nothing new and it works;)
I guess Google does say that "Simply disavowing them (spammy links) isn't enough."
Can you (Josh) speculate on what is happening when a person disavows links, but it isn't "enough"? You suggested in a negative way that people might be disavowing helpful links. However, it seems Google is implying something else malicious that is occurring when we merely use the disavow tool. Any comment?
It's certainly an interesting topic for discussion. Although the theory of is a disavow file enough to get out of a penalty may or may not be true I certainly wouldn't be putting all my eggs in one basket. The disavow tool was designed so when a Webmaster was unsuccessful in contacting and manually removing spammy/bad link it could be discounted, it's basically the last resort you're meant to take. I'd personally be both contacting website owners and disavowing the bad links at the same time.
I'll agree with the not selling a link disavow as a singular service unless it was combined removal of links as well. Ideally though it should be a part of a holistic marketing campaign as there are other ways to improve rankings and traffic whilst also trying to recover from a penalty (it's not the end of the world anymore).
Disavow tool is a great way to aid recovery from a penalty as long as it is used wisely.And It is of the utmost importance that one can cover every square inch of backlink profile to stand the best chance of recovering from a penalty.
I think It's pretty safe to say that the disavow alone does not get you out of a penguin hit. The keyword there is alone. Out of all of the sites I have helped with, I commonly find that they have had some old school web spam links somewhere in the link profile.
I tend to disavow these links when they are obviously spam. I don't know that it definitely helps but It's been a part of my process for a while and I have been getting results this way. I have never found a site which has seen a dip in rankings and the only problem has been spammy links. There are always other problems which need to be worked on.
I guess it's food for thought, am I adding an unnecessary step to my process and approaching this with an "it's not broken don't fix it" mentality?
Im not Joshs biggest fan but i have to agree with Josh on this, i dont believe the disavow works. I think its nothing more than a propoganda tool, and the only real use it has is for GOOGLE'S benefit NOT for the guy on the street. I wont go into detail but we can all imagine scenarios out there why/how Google could use the information handed to them in disavow files. In fact, although it pains me to say it the idea of a disavow file is genius- it gets people to effectively show Google where all the "sold/spammy" links have come from.
Ive seen comments by those saying theyve done X amount of disavows and theyve worked, if that was true then it would be more common knowledge that it worked but from what ive seen - and i mix with some of the worlds best marketers every day - on the whole it doesnt work.
What i think we are seeing is that there is a deliberate randomness thrown into these types of things placed there by Google to create uncertainty, and most importantly FEAR. If some people use disavow and seem to recover its gets shouted bout and enough people will assume it works, if not then we have to be afraid of the bogey man watching our sites. If they instill enough fear into things then people run around like headless chickens and the only safe option is to go to paid search..ie adwords. Also it just seems obvious to me that to go and actually take care of these 10's of 1000's of disavows is not cost friendly for them so why would they?
The amount of people on here who have commented who are naive enough to think that Google actually care about if you need your site to recover or not is amazing. Do you people really believe that if you/ your clients get a load of crappy links then Google are going to help you get out of the mire? get real. Josh is right - they DONT CARE. The sooner you realise that the better.
Hello Joana The Disavaw tool works when you receive a manual penalty. There will always be links sitting on websites owned by webmasters who have either jumped ship or don't care that are almost impossible to remove. Google seems to understand this. Without the Disavow tool I personally wouldn't of had my Google Manual Penalty revoked when I did
Hey Alex, im aware of what the disavow does/claims to. My point is why do you or anyone here for that matter think that Google care enough about you to help your site recover. in their eyes, someone has done something either they shouldnt have or "naughty". Why would they then care about helping you correct that, this would cost them time and money- they are a business, it doesnt matter to them that you had spammy links. When have they EVER done anything for any of us? unless its for PR of course.
I think the only sites that seem to recover are just nay random sites as without those random few it would be obvious that the disavow didnt work at all and that would warrant unwanted attention from people. Its just business to them.
I didn't say google cares. However, they helped me remove my manual penalty by letting me know what sort of links that they deemed "spammy/low quality". Someone in the manual web team was being paid to do that, they didn't have to go through my site.
However, we seem to be missing the point that Josh is making. Can you recover just by submitting a disavow tool. I agree with that, no.
Hey,
I tell you what i did after 17th OCT hit. First i identified the risky links and then i removed as many as possible over next 4 weeks and then i disavowed the remaining. However, the demoted a dozen websites didn't returned to first page until i removed on-page keyword stuffing and not until removing links to external websites. Once i did the onpage cleanup, i discovered the sites on page one again almost overnight. Read my blog on my profile to read full story.
I personally enjoyed disavow tool my belief on this is disavowing is not only the solution for Penalize of website, manual removal is also the best approach to recovery of Penalize website you have to perform all the process of links removal, disavowing, documenting the link removal and sending recover request.
Unfortunately I think this experiment is flawed from the outset, especially where penguin 3.0 is concerned. It has been almost one year since the last penguin refresh in which time spammy links will be deleted and created. I also believe that Google looks at the last time the algo was run and not at individual time stamps when disavow files are submitted to determine whether a website has fallen foul to Penguin. Maybe in future we will get a better picture on whether the disavow tool works now that we are told by Google that the refreshes will be more frequent allowing more chance of your experiment to work. I do however agree that the best action is to remove as many links as possible before disavowing them. I also don't believe Google will simply recover sites from Penguin purely based on what links are in the disavow file.
WBF is awesome, Got some good tips.. Thanks Josh Bachynski ....
The Ultimate Guide to Disavow the spammy links here.. https://moz.com/blog/ultimate-guide-to-google-penalty-removal ......But Tryed for few of my sites, I am not sure this will work because still I am not getting the result..after a year.....But the process is awesome and for that need time.......Thanks...
The fact of the matter is that all of the information we have is anecdotal. We all have experiences, and only Google knows the answer as to what works and what does not. That being said, it makes sense to cover all of the bases.
Josh I think No one can give you a perfect example that he or she got benefited just by uploading a disavow file. My experience is a spam link should be deleted and also should not be found by in the next crawl. Need to check the last cached page, if the link is showing in the last cashed page it means the link is still having bad impact.
Disavowing a link may help if anyone tried to remove any link but failed. Not sure it helps or not. No proof .. :(
You claim to be "scientific" but when you say "link loss" you never mentioned how big was the link loss - 1%? 10%? 90%? And - more important - what was the "quality" of the lost links?
Penguin hits websites for big amounts of artificial anchor-rich spammy link profile. If the "link loss" is just 1-3% drop - it cannot recover you from Penguin, since a 1-3% cannot be a reason for a Penguin hit even if the links suck.
Actually, it is very easy to see, if the link loss was the reason for Penguin recovery - by looking at the lost links.
Hi, Excuisme, I am From Indonesia, Thanks This Articles very help me..
Best Regards
MutiaraPublic