I wanted to give folks an update since there were a number of questions regarding Google at the end of September.
We found a significant regressions that in early September let a large number of old profiles into Google's index (Millions of pages became indexable). We put out a fix as soon as we found it and expect that over the next 60 days or so, Google will digest the change. Google never likes getting millions of low quality thin pages:(
The good news is it's fixed:)
We found a few more things that we think could be better and have fixed or will release fixes this week.
We bounced the situation off a few of our favorite resources and think that the fix is a good one and that it's just a little wait and see now.
Back to work....
Nice job Paul! I knew for sure from the data that I had that it was a low-quality page issue, and obviously, the only thing that was visible to me directly was the QnA feature and so I apologise for the false accusation there. I still am not an absolute fan of that feature, it could be a lot better (but the idea of qna, on the whole, is amazing), but I sure am glad you figured out the real issue.
Yup, we just got to wait for Google to deindex those pages. I'm really glad to see that you posted this, it does help bring back some lost confidence.
I'm sure many of us just heaved a huge sigh of relief. Many were very upset at the huge losses of traffic. This is especially bad since this is the holiday season and many here count on pageviews and sales during this time to earn them money. Thank you so much for posting this.
Good to have a positive response - a relief for many I'm sure - could be a very good Christmas and New year.
Good to know. Thanks for your hard work and letting us know what's happening.
Many thanks Paul and your team. I realized visitor traffic on HP had been much less than before, which got me worried. Now I know and really appreciate the effort that your team has put in. I understand the challenges you and your team face since other publishing sites have closed. HP is among the few survivors. It survived because HP has a great proactive team. I'm sticking with HP for good.
Wow, good to know you found the reason for the traffic drop and corrected it. Thanks also for the transparency Paul.
Thanks Sir. You folks are so on top of things, always. I only hope I can create stuff worthy of being a part of the network.
Quick thought if you've got the time to answer: Shouldn't these profiles that got indexed be on the HP domain and not on the niche sites?
The articles I rememeber where definitely on the niche sites or the new site Maven acquired. I wonder if the issue was partly caused by the recommended section that they have been experimenting with?
I know you never saw that, but I'm still seeing it and it has greatly improved and the recommendations are now mostly related to the articles I'm reading. With the exception of Owlcation articles. Those are still all over the place, but at least they stay on Owlcation (which wasn't always the case when the feature first came out).
Glad to see the issues were caught and are being fixed. I'm still concerned that it appears hp is not truly separate and can greatly affect the niche sites.
I just checked on the uni computer and the recommended articles are good. Owlcation is as you said still not directly related, but it's showing Owlcation articles. Thanks for pointing that out.
I am still seeing it. It has NOT improved. The same MTO articles on pubic hair, "Wow did she get those breast implants TOO big?", etc.
Just checked your article: treat your dogs yeast infection at home. The recommend section looks good and some of your hubs are listed there. Maybe it's some sort of caching thing that prevents me from seeing recommend articles on my laptop that is causing you to see the old links? Hopefully...
The pubic hair article I was seeing was an hp article not an MTO one. I was searching this morning and did not see any. I wonder why some people are still not seeing improvements. Mine has definitely improved from when I first started seeing this section.
Are you seeing it on pethelpful? I didn't really check that site as I could not remember seeing them heavily there. But I did look through a few of my articles and didn't see anything concerning in the recommended section last night or this morning.
That would certainly be a disaster on Owlcation. I can imagine a lot of parents hitting the back button fast if they are helping kids with homework and silicone implants start bobbing around (hope we are not talking animated gifs, lol).
Google notes that kind of user behavior.
Some parents might even block the site in safe search.
https://www.google.com/webmasters/tools … &pli=1
That is not a good signal to be sending.
I think I saw it on the HP main site. The child safe site, right?
I don't think there is anything intrinsically wrong with adult orientated pages or ads as long they are not offered on sites that kids regularly visit.
I think most of us that perceive the MTO ads (or recommended hubs) as a problem do not like them because we have been told numerous times that the HP site is clean. Even the "Indian aunty" type pages with their mildly sexy poses have been removed over the last couple of years.
(Heaven forbid that we should put anything graphic on the site. A dog having surgery? No way, too graphic.)
Thanks for confirming what we already knew. Glad to hear that you are doing something to address the situation.
Thanks for letting us know. It's always good to have reasons behind these things so that we know what is going on!
Thank you so much for sharing this information, Paul. I appreciate learning about what has happened and about what is being done.
Sounds as if the niche sites are vulnerable to many outside influences. Or were these profiles on the niche sites themselves?
Excellent news that the issue was found and corrected quickly. I'm glad to see some red arrows again!
Corrected...mostly...red arrows? it's going to be awhile. Paul says perhaps 60 days until Google catches up with the corrections. Till then, it will be low views and low money for us...which is really bad considering that we are entering the holiday season when many here usually earn well.
It's good to get some info from you about what's been going on Paul.
I don't think anyone could have predicted this particular scenario was the cause of the traffic changes.
I hope the traffic declines reverse real soon now that you've fixed things.
Paul, Thank you for the update. I am happy a fix is in place.
Thanks for letting us know, Paul. It's good to know that you guys are on top of things and always looking for what needs to be addressed on our behalf.
It's always worth trying to learn from disasters.
If panda issues on hubpages.com hit the niche sites, will YMYL issues on HP have similar knock on effects? Or penguin issues for that matter.
Also, is there any real benefit for the niche sites in being linked to HP? Could we all see a big traffic boost from complete separation?
Writing on the wall, shouting at your mum, sucking your thumb...
I get ridiculous amounts of criticism for those kinds of things.
When your hubs get sent to the naughty corner.
It gets even worse when your GSC Data is is wonky, Sherry.
But seriously, I interpret it as meaning that lots of crappy hubs that HP have worked hard to exclude from Google has suddenly come back into play.
So, in other words HP screwed up? Is that's what he's obfuscating? Hey, we all do it.
Not sure it is obfuscation. I think that is the way people in California speak.
edit: I've lost around twenty percent of my income over the last couple of months, so I reckon staff deserve at least a 20 percent roasting for the next 60 days, lol.
I lived in California for 45 years. That's not how I speak
All that really matters is whether it's a technical matter that HP need to work on, or whether it's a matter for writers to deal with, editing etc. It sounds very much like the former. So I can pour a drink, put my feet up, and maybe write another hub.
There is a lot going on with search right now. My feeling is Google has a processing issue at the moment. Data in GSC is wonky, buggy behavior for site: searches and several other peculiar things I’ve seen. These things usually clear up, but we will hit a few of our contacts to make sure they have the info.
I've got to wonder if a Maven tech glitch was tangled up with this in any way. The timing of the dates is coincidental to say the least. https://maven.io/the-maven/investors/ma … kGGVAZIjA/ As a former programmer/analyst, I can indeed attest that tech is a pain in the neck.
Thanks for the update, Paul. It sounds like a technical issue that individual writers have no control over. My earnings have recovered to some degree. Hopefully, it will be onwards and upwards now.
Paul spoke in regular tech terms, not Californian, but English
So it is a worldwide assault on the English language?
We are still referring to significant regression right?
Definition: a return to a former or less developed state
Former pages (profiles he says) were re-indexed and the result is a significant loss in traffic as experienced before. So this is a significant regression and not a complete regression.
Unless y'all refer to something else?
This is plain English:
Someone messed up with the coding and it cost everyone associated with HP a lot of money.
I disagree. If he just said that we'd all have a lot of questions and it's very clear that he does not have the time to respond to every question. It is a lot easier for him to give us the facts and then if someone has questions, they can ask them. There are other hubbers who can help, as you and a few others have done here or it is possible to use a simple dictionary. When dealing with any field, the only way to get the right and complete message across is by using the relevant terms.
Sorry, Brandon but this does not make sense. It would have been easier to simply explain what actually happened in the original post so that none of us would be forced to waste time speculating.
He could also add the plain English thing saying it was a mistake. But it is important to know what the mistake was.
All that was needed was this: a large number of old profiles into Google's index (Millions of pages became indexable). We put out a fix as soon as we found it and expect that over the next 60 days or so.
If this part was not in the message and Paul just said we made a mistake and we rectified it, it helps to know they fixed something, but what and how? There are a few who like to keep the technical stuff as a black box, but there are others who want to know what's happening.
If we have just the black box answer, it's useless to some of us, but if we get the tech answer, there are others who can decipher it. That's my take on these things.
So, this "regression," could it have happened all by itself? Could the old profiles just seep out somehow? I am not trying to assign blame. I'm sure HP was just as negatively affected by all this as we were. If it was a mistake on their part, I'm sure it was an accident. However, it's probably not something they want to advertise.
I still don't get it. I can see articles that had 100 scores dropped so low they had no traffic, and I am still catching up. I've recovered maybe one quarter from where I was, and I was at the top of my earnings after all my years here, so it really hurt me.
Yup, this year was the best for most if not all of us. My traffic is still recovering too. Some others have picked up, but the old time best hubs are down a substantial amount.
My traffic has dropped about 50% with minimal increases on weekends, as has been the pattern. It seems to have become stabilized at that level, which is not anywhere as good as it has been. My only good news is that if Paul is correct and in about 60 days the problem will be fixed, it will be just about time for my season to start again, so I should be able to recover. However, if this is something more ongoing, it appears to me that while I'll still be earning and getting views, it won't be anything like before. Also, my Adsense income has dropped now down to pennies a day, so this thing is affecting income from that, too.
Deepest thanks Will because I had no idea what all the previous chatter was about.
Well, I'm calling this episode Codegate. We may or may not recover. And we will never truly know why it went one way or the other.
It's complicated though, close to what Americans would call a game of "whack-a-mole". Just as Google can change it's algo slightly and cause ripples of unintended or unexpected consequences, HP has to keep adjusting, just as a traveler has to keep adapting to the ever-changing weather. I am mixing my metaphors here, but generally HP has been good at adaption, so they have earned my confidence. Look at all those who fell by the wayside over the years. It's been a rollercoaster ride for me, but overall the direction for me has been upwards (okay, that's another terrible metaphor, given that rollercoasters rely to a large extent on gravity!).
Suffice to say, things seem to be getting better for me after a dip. Hub Ads anyway, Amazon earnings are as a crappy as ever, but maybe that is the way it's going.
What I see is a huge amount of work and effort from staff and writers let down by someone who does not know how to tag a page properly. It is infuriating and very hard to excuse.
This might be the reason:
https://support.google.com/webmasters/a … 3710?hl=en
If not, it is something similar.
There was discussion quite awhile ago about the fact that HP does not give us a way to use meta tags on our own articles. Could this be something that would fix this problem? How would it be implemented if doing so would work. Would this not fix the problem you just mentioned? Also, why would anybody want to put a no index tag on their work? It seems to me that if work is not indexed it cannot be searched. Am I missing something here?
HP allowed all kinds of very poor quality pages to be published at one time. Back then it was an advantage to have very large sites even if the pages were worthless to readers.
Content Farm sites like Ezinearticles, Squidoo and About.com all did well in search but Google got tired of offering poor quality results to searchers and cracked down. Most content farms folded.
HP survived, in part by defeaturing/no indexing the worst pages, improving quality on the remainder and putting the very best on niche sites.
Unfortunately, the worst pages (that were supposed to be invisible to Google) became visible a couple of months ago. Someone made a mistake in the tags.
The element in Google's algo called Panda immediately identified this huge pile of cr*p on the site and gave everything associated with HP a penalty.
Thus we have seen a big traffic drop and earnings drop.
We don't need control of meta tags, we just need someone who knows what they are doing to be in charge of them.
Having worked as a programmer, I feel like I know enough about coding and computing to understand how little I know. Tags, links, etc. are simple in themselves, but put together on a huge scale in conjunction with the most complex algo ever, you get an unstable relationship that takes skill to handle. The Google algo can only be really be guessed at through inductive reasoning. It's a game of cat and mouse. HP has played it well over time, even if there have been some major problems at times. This latest issue also seems relatively minor compared to some previous falls when traffic went down 95% not 15 or 20.
But as always, the proof is in the pudding.
Ten out of ten for loyalty but facts are facts. SEO cockups equal traffic misery.
Whatever caused those second rate pages to seep back into the SERP's was avoidable or it would not have been fixable.
Better to dump the garbage pages completely if you cannot keep control of them.
So why in the world didn't they? Who keeps bad work available once it has been removed? Makes no sense.
All the non-featured hubs have a no-index tag, could be the no-index became index. There are many non-featured hubs, and they are not deleted. This is why they are not removed because people can still access them through the direct URL.
A lot of people have put a lot of work into writing for HP. Some of them are not very good writers and never will be, sadly. Some of them will learn given the encouragement.
HP took to unfeaturing pages (hiding them from search engines) partly, I reckon, to give newer writers a chance to improve without the slap in the face of deletion and partly to avoid outcry from regulars.
Unfeaturing is fine as long it works.
I may be completely wrong, or just plain reductionist, but I suspect that there are technical advantages to sheer bulk of pages being linked together. HP are trying to strike a balance between harnessing the power of the links without losing kudos for thin content. When Google f**k around, the balance can be lost, but then regained, Grasshopper.
All these cr*p pages are on the HP domain, I still do not understand why the niche sites were affected. Each niche site operates on its own. And I doubt there were ever so many cr*p pages on the niche sites which are now set as no-index.
The niche sites clearly do not operate on their own. This is something that we have just learnt.
Exactly. This is the bigger problem here, not the problem that some old pages got indexed. That's a big deal too, but the fact that they do not operate on their own is the issue. It's also a question on my second reply on this thread. Paul did not address this issue.
I suggested earlier that it might be worth separating the niche sites entirely from hubpages.com. I reckon they should consider it especially for the YMYL sites like pethelpful.
As far as I'm aware, nobody knows what kinds of signals the search algos use to establish EAT (expertise, authority, trust) but links (guilt by association) might well be involved.
At least make more use of nofollow links so that the HP network isn't one big boiling soup of very patchy quality.
edit: That last sentence is rather weird. Try to imagine big bits of gristle floating next to tasty wild mushrooms.
Frankly speaking, the person who made this mistake should be fired. There is no excuse for this type of error. Yes, people are human, but I'm sure the team pays their tech people good money for their "expertise", but when someone lets this type of blunder occur, everybody suffers and it puts the site in grave danger. What I don't get is that if someone simply pressed the wrong button or applied a bad code, why is it so hard for another person to undo that button or code. This does not just affect one individual, but thousands of writers and articles..I don't understand how something so wrong could have been done and why it will take so long to fix the problem.
Oh my. This can be caused by the simplest of errors and is definitely no reason to fire someone good at their job. It's not a button but some code that went wrong. Not sure how it works on HPs end, but when I work on my client websites, usually wordpress websites all it takes is the deletion or an error of one line of code to cause this issue.
It is a simple fix once realized, but those pages were already discovered by Google and now we need to wait for the Google bots to come back to those pages so that the bot realizes that they are set to no-index and hence take it off the Google database. This is why Paul estimated 60 days. Got to wait for Google to get back to those pages.
No doubt it is an expensive mistake, but finding another person and getting the new guy up to mark with the system would take longer. Also, why fire someone good at their job. He or she would now double check this everytime, a new person would very likely make the same blunder sometime in the future.
Totally agree. I'm sure most of us have made mistakes professionally. It's human nature. The only reason to fire someone would be if they're constantly making mistakes and incompetent or totally irresponsible.
I see your point, but if was a "simple mistake", then the fix should be simple as well...which it obviously is not. I like Kenna McHugh's comment about the importance of supervision for people who do this kind of work. It is when that is lacking that these things happen. Perhaps the team should hire a few SEO specialists to oversee the work that our programmers do so they can avoid this type of problem in the future.
A stronger surge of traffic this morning via Google on my Owlcation articles which had recently bottomed out to zilch ... looking better ... promising. Love to see the red.
I've worked with software company developers. The way you catch code mistakes is with Quality Assurance Department. That is the sure fire way to catch mistakes because they do happen. Just like a writer has editors.
Just recently became able to post on the forums again but my two cents worth is that HP will handle the problem as well as they can. Paul has been able to save this site before when things looked grim and he'll do his best now. I've learned to be patient during these minor crises over the last decade with HP and I'm not panicking now.
Well, I don't quite understand Paul's update but I'm hoping it means there's a reason why my daily dollar numbers have fallen drastically, and that there's a fix in the works to make them more palatable.
Should we be seeing an improvement in traffic yet due to this correction?
(or will it take at least 60 days for a significant improvement?).
If it's working you should already be seeing changes happening. It's not going to be on one day where it goes boom.
Overall, my views are slightly up, but not on the hubs that lost the most traffic. The ones that went down, stayed down.
I get two or at most three days at a time that are up, one or two days from Sept to now that are way up and the other days where they were or down further. Earnings are slightly up but are still averaging out to what they have been since I started making payout monthly. Still seems that when my views go up my CPM goes down and vice versa so no matter what every month I make about $55 to $62. It remains the same no matter how many hubs I have written or how many views I get such that I sometimes can't help but think it is being artificially controlled.
I used to think that higher CPM meant lower views but recently this theory has been blown out the water. CPM reached an all time high, easily double the usual rate, alongside stable views. So I don't know what's going on! I'm glad you're reaching pay out, that's your reward for hard work...perhaps more quality articles are called for, build up momentum going forward.
Today has been my best day to date traffic wise. Most of my traffic has recovered, but until today I haven't seen any rises in traffic past what I was seeing before the sharp decline.
I was starting to think I would never see over 300 views for some of my articles that hold the #1 spot on Google, but I might just be proven wrong before the year is over. Have been reading up on SEO and trying some of the tips that Brandon has been sharing over the last few months and I think I might finally start seeing some increase from the work I've been putting in.
Awesome! Online publishing is the name, patience is the game.
Good news. I think Brandon might be looking for a percentage if we're not careful. Seriously, keep at it. Build up momentum with more quality articles, persist.
I'm still not seeing a recovery from the big drops we had a couple of months ago :-(
I think it's Owlcation and product hubs that have done particularly badly, Will. That's my experience anyway. But things are fine in other fields. My earnings overall are good. I'm kind of glad that I cover a variety of topics, more by accident than any deliberate policy.
Seems to me that the losses are anywhere that authority matters. Those "millions" of second rate pages that flooded Google are like bloated corpses bobbing to the surface of a lake, guilty secrets that HP should have made sure were gone forever.
You spin a macabre image. My views are reasonable and my earnings are the best ever. But yes, it would be nice if views were back up to peak levels across the board. I've pretty much given up on Owlcation for the moment - and Owlcation and product hubs were my thing for some time.
It would be nice to get another HP update on the "fixes". I suspect they've only been partially successful...
I'm expecting, after the holiday season, that January will be kind to those of us with non-holiday related hubs. All we can do is wait and see.
Well, if Paul was correct about it taking 60 days for Google to "catch up" with the issues, January would be the month when they do.
Boy did I get that wrong. Traffic fine and dandy; but those pesky, cheap advertisers...
It's obviously not a switch or a tap that makes traffic flow, so don't expect spikes in traffic in January. Any benefits/recovery would have been happening from the day they made the fix.
Will benefits continue though or would they have plateaued by now?
It depends on whether those pages have all been out of the index or not. I can't say, but from experience the pages are not out of the index so quickly. But I wouldn't expect to see any further benefits from this change, because any major benefits would have been seen in the first few months. Traffic has in fact risen (not as good as they were in the summer) a bit across the niche sites based on SEMRush reports and my own stats.
I guess I am lucky that I had one of my properties (accounts) moved to mobile first indexing right in the middle of all of this. Smartphone Googlebot crawl rate goes up. Seems like it gets the index moving more quickly.
What do you mean by "one of your properties"? If it's one of your own websites unless you had a huge blunder that should not really make things any different. Also, Google now has mobile first indexing on all websites so HP should be the same.
"As we said, we transition sites slowly to ensure a good experience for site owners and users. We evaluate each site individually on its readiness for mobile-first indexing based on the best practices and transition the site when the site is ready."
https://developers.google.com/search/mo … t-indexing
Google emailed me a few weeks ago and told me my hubpages property was moved. When you get moved you get and increased crawl rate which leads to extra trafic. Its a transition. It was just my turn. It was my we have kids account.
Hmm. These notifications are coming from the Google Webmaster Central. How do you have a hubpages property claimed? We cannot do this anymore as far as I know. The Google Analytics verification method does not work. The sub-domains were the last time we could claim and keep track via Webmaster Tools (search console).
If you do know of a way, it would help many of us here who would love to have it back
I dont have a verified property in Webmaster tools as far as I know. Property was the term they used. The whole thing confused me too. I cant see the crawl rate, but I see my traffic on that account bouncing all over. It's up overall, but it's a small account right now. I don't know why the notification came the way it did. It's interesting though, and I think the movement of mobile first indexing is something to watch.
I had the same notification. It's a generic (not individual account) note. Go to your Google Analytics page. In the menu at top left is a bell symbol. Click on this and it will list all the Analytics Notifications that impact on your account.
I've been waiting for the fix to kick in, and it didn't. Not for me. So, from what I gather, it's not going to happen. Well, again, not for me. So, it's either my articles themselves are what have dropped or the fix didn't work.
HP might like to look at every change they have made over the last 12 months. A short list:
Inappropriate and unrelated "related pages" in the side bar.
Do follow links to and from Maven
Every article they have added
There is a lot more, I reckon.
Maybe dial it all back.
Yup, all of these things that you mentioned are also leading to a horrible user experience.
Seems as if HP and some niche sites have taken another hit in the last few days. Traffic is down, once again. The perpetual battle goes on.
Yes, I noticed that last week my traffic cut almost in half on my other profile. (I have more than 1 profile due the transfer of Squidoo articles in).
I had the worst traffic this month that I've had in a year or more. But December was the highest ever. It's so confusing.
I am also having trouble with Owlcation, I submitted an article over 6 weeks ago and nobody has contacted me. Either take it or leave it, but it takes less than 5 minutes to read the average article, so I can't believe these editors are overwhelmed with stacks of material. Many aren't even writing new material. I just began again after two years of "fixing" hubs to get them up to standard, and now wonder if it was time wasted.
When an article is removed from search does it receive the half circle or does the writer have no way of knowing his article has been hidden from search?
When you see half-shaded circle, it means, the article which was featured earlier is not featured now.These article move from featured to Non-Featured state because of Traffic.
I am not sure what traffic threshold number from google makes HubPages to decide an Article is Not Featured. Note, the half-shaded circle does not complain about Quality but Traffic.
The advice is add one or more section(s) to article. I had done that for few of my early articles (Which is not in niche anyhow) and got back featured. May be other users who had experienced this shed more light on this.
The threshold for traffic from Google is VERY low. A few years ago Paul stated that an article only needs to have a "heartbeat". If your article is seeing any traffic from Google, it is not going to be unfeatured. If your traffic is only coming from social media, and from your shares, it may be unfeatured, but it will still show up when someone from a social media site is looking for it.
If you add a new section, it is my understanding that the article will be searched by the Google bot. Once again your article will have a heartbeat. If you do not get any traffic after your edit, however, it will become unfeatured once again. (The amount of time your article has to be unfeatured varies by the hubber score. I think everyone has at least a year.)
Copyright © 2019 HubPages Inc. and respective owners. Other product and company names shown may be trademarks of their respective owners. HubPages® is a registered Service Mark of HubPages, Inc. HubPages and Hubbers (authors) may earn revenue on this page based on affiliate relationships and advertisements with partners including Amazon, Google, and others.
|HubPages Device ID||This is used to identify particular browsers or devices when the access the service, and is used for security reasons.|
|Login||This is necessary to sign in to the HubPages Service.|
|HubPages Traffic Pixel||This is used to collect data on traffic to articles and other pages on our site. Unless you are signed in to a HubPages account, all personally identifiable information is anonymized.|
|Remarketing Pixels||We may use remarketing pixels from advertising networks such as Google AdWords, Bing Ads, and Facebook in order to advertise the HubPages Service to people that have visited our sites.|
|Conversion Tracking Pixels||We may use conversion tracking pixels from advertising networks such as Google AdWords, Bing Ads, and Facebook in order to identify when an advertisement has successfully resulted in the desired action, such as signing up for the HubPages Service or publishing an article on the HubPages Service.|