Thanks for sharing this article. I had missed this one. It was a real eye opener.
Thank you for sharing this. It is good to read what's on the minds of the people behind HubPages. From reading the article, it sounds like a lot is going on in GoogleLand and HubPages is working hard to stay up with Google to make things better for us. That was a good read.
Thanks for sharing it, paradigmsearch! Perhaps it can provide some helpful history and context to new Hubbers on the site (plus reaffirm our major goals now- something relevant to all Hubbers).
It was also nice of Paul to give Mark Ewbie a shout out like that.
I had the fortune of encountering Mark Ewbie in a forum discussion just a while ago.
I find it interesting that "as the quality of the page increases, its effective revenue decreases". If your page is very good and informative, there is no need for the punters to click on ads, they get their info from you.
Which is why HP's pay-per-view ad program is such a boon. Nobody clicks on the ads on my blog!
Like that thing they mentioned yesterday about Maholo (I think) in yesterday's piece, that their instructions for authors told them to write their pages so that the ads provided the info the reader was looking for, not the content.
It's also probably why writing Amazon pages works better for me in terms of earnings. With review pages, decent content encourages people to buy the products, so I suspect the revenue increases with content quality.
Am I the only one to find this statement ironic:
"I was a bit surprised that they would opt to utilize a brute force tactic as opposed to taking a more surgical approach in which low-quality pages would be singled out and given lower rankings. We felt Google’s actions communicated a sentiment that content is expendable on a massive scale"
Is this not EXACTLY what HP is now doing by punishing authors whose only failing is that they have not mastered (or lack time to implement or money to pay others to implement) the tricks for gaining traffic?
The idling of hubs because of low traffic is a brute force tactic, but I guess HP would argue that since hubs to be idled don't get much traffic anyway idling them doesn't make much difference to revenue, whereas Panda cut the traffic to good quality pages which were making a lot of money.
I actually found it very surprising that the whole idling/featured program wasn't mentioned (except to say that tag pages and some questions were idled). Isn't this the biggest thing that is happening on HP right now?
You can't really say that it is to protect commercial secrets, there is plenty of info publicly available on idling hubs.
I was also hoping for some discussion of whether the subdomains strategy was ultimately a success, it seemed to work initially, but then 'SEO visibility' plummeted.
Yeah, that seemed like a real puzzling omission to me. I have to wonder how the site is doing a few months after this has been implemented. I've been ambivalent about it myself. Since idling (excuse me, featuring) started, I've seen a pretty significant return in traffic. It's still about 50% off of what it once was, but that's a damn sight better than the 90% I was living with.
The uncertainty is kind of frustrating. Were my trash Hubs bringing me down? Is it just luck? Would it have happened on its own as Panda rankings slowly readjusted? I was really hoping for a more insightful discussion into the mechanics and results of the new initiative .
Welcome to the internet. Where it takes a quick action with a quick response to be able to assign cause and effect.
We know Panda hurt, we know subdomains helped - both action and response was nearly instant in both cases. But we will never truly know if small actions carried out over a year or more has helped or hurt, or whether continued updates by G was actually the cause of traffic changes during the period. Even if traffic falls over the next year, for instance, did the idling process limit the fall to less than it might have been? Or make it worse?
Re: subdomains- they did make a big difference, but Google has come to look at authorship differently over time. So it was the right choice at the time, but given subsequent algo updates, we had (and have) to do more.
No you're not the only one. I thought exactly the same thing.
Hey WriteAngled- we see our QAP as being (as much as possible) a targeted approach. Targeting Hubs based on traffic (in addition to human quality ratings) has enabled us to move faster, which is important. Yes, it would be better if we could have just run quality assessments, but we don't have the resources required to do that in a timely manner.
Paul didn't refer to the QAP and Featured Hubs in the article by name, but did do so through his broader (more HubPages-ignorant-reader-friendly) description of what we're doing.
Yes, you are targeting and crucifying quality writers who have not mastered dirty little SEO tricks and do not have the money or inclination to pay others to implement same for them. You are censoring them out of the search engines on the back of one statistic, namely traffic. How many good writers have said on the forums here that once a hub is idled, they simply move it elsewhere, because they are not willing to play the endless game of it going in and out of the search engines?
You are targeting and crucifying quality writers who choose not to deal with what the must-have shoe colour is this season, the newest electronic gadget, how to make a meal in 5 minutes by opening two tins, the latest celebrity gossip and all the other pap that is loved by the masses.
You certainly are moving faster in thoroughly demoralising and demotivating many good writers. Meanwhile, you are happy that the stinking garbage of Indian aunties and diverse vacuous content in broken or spun English continues to fester at the bottom of the pile, where it has been rotting and pulling the site down for the last five years
I see absolutely no point in adding any more content here and then have to go through the constant stress of wondering when the no-index tag is going to fall, and knowing that whatever I do to revive the hub, the axe will fall yet again, and again, and again.
What 'dirty little tricks' are you thinking of? I'm not sure that any still work. Only a few writers here ever bothered with them, when they did.
As for the shoes and gadgets, a lot of people genuinely care about those things. Capitalism would collapse without consumer addictions. I rather like HDTV's, cute little tablets, Kindles, etc, myself.
Have you ever thought that you might be better off contributing to Wikipedia?
If the answer to that question is that 'I would rather earn some money' then greasing the wheels of commerce is the best solution.
I would sooner commit hara-kiri.
I consider Wikipedia one of the worst phenomena on the Internet and never trust a single word written in there since there is no way I can verify that the contributor is an authority on the topic in question.
Veracity cannot be decided by anonymous committees such as rule at Wikipedia.
Paul's interview seemed to indicate what many of us suspected, that shorter hubs trumps longer quality content. This may be why those very short Ehow articles top the search listings in many categories. Not good news for those of us who create well researched and detailed content. It also helps explain why so much dross is getting by the vetting process. It's what HP wants now, it seems.
I found this very interesting:
"we heard from Google that one of our authors (who we thought was pretty good) had quality issues"
Shouldn't Google provide some system where an author can check their quality? They have this whole authoriship thing - so surely they can apply a 'score' - it seems to be very important to them!!! I'd like to think that I am a good writer - but if Google disagrees I need to know this! Not sure if this is information that Hubpages has though?
That is the exact line that caught my attention. I would be curious to know, really curious, exactly what types of quality issues Google detected.
@Randy Godwin - Crappy content can out earn good content because the ads are better than the content, but nobody is interested in sending traffic to those pages so they're effectively worth very little.
"those of us who create well researched and detailed content" will ultimately get the lions share of traffic, but the pageviews will have lower effective earning rates.
There are a couple of deflationary aspects going on. First, the switch to mobile. Mobile monetizes lower than the web. I expect half the traffic to be from mobile devices in the US by year end. The other is as people compete for the top spot in the search results longer more detailed pages will earn less than short, mediocre/crappy pages.
Roughly 39% of my traffic seems to be from mobile technology. One thing I've noticed from my data, as well, is that the visit duration from mobile devices is 57% of what it is for non-mobile devices. This makes sense, but it makes me wonder how I can design content to appeal to both types of audiences (mobile vs non-mobile) better. I'll have to think about this.
A very good question, and one that I've struggled with. Detailed "how to" instructions (full of photos) are not particularly conducive to mobile devices and that's mostly what I do. A problem I need to continue working on even though the percentage of my mobile traffic is not that high yet. Yet - I'm sure it will grow.
@SimeyC I wish Google were transparent about quality. They hide behind that if they make it public, people will use it to game them. That type of thinking really hurts people that want to do the right thing. We are collecting more quality information on authors, but we don't have enough yet to be all that useful to authors.
However, we are getting close to being able to compare the quality of the HubPages corpus to other large sites. Many people feel that quality is all about thresholds. Get on the wrong side and get wacked. I'm hoping we will have useful data to share in the coming weeks and months.
They have to keep that stuff secret. For every 1 person who wants to know to write better, 99 want that info to take advantage of the system and make as much as possible with as little effort. That's how content farms and backlink selling got started in the first place.
There are two ways to rank higher, using tricks to make the search engines think your content is more popular than it is, or time. Your policies make the second way impossible. My account is only 6 or 7 months old, I shouldn't have a single article that was de-indexed. Yet almost 70% of them have been, most done within a month of writing them.
Fresh content is what ranks, not deleting stuff. That's the reason my blogs have all gone up in rankings while my HP continues to go down, despite my HP being older than all but one. With pretty much the same amount of effort, any one of my blogs will get more traffic in a day than my HP will in a month.
I have seen the staff say that they are looking at this process and extending the time before idling. From everything I am learning, it does seem like it takes time for good content to climb the search engine ranks. Hopefully this will happen soon!
I can't believe Google even try to make any kind of qualitative judgements about quality. They probably settle for user metrics except, perhaps, for their limited sandbox tests on new search algos.
This is why I reckon the mTurk check is probably only useful for weeding out the very worst stuff. It won't really tell you if a page is going to satisfy a reader who is genuinely interested in the subject and maybe pretty knowledgeable as well.
Why doesn't HP do the simple thing and idle pages with pitiful read times? That should satisfy Google more than any other thing.
Great interview, Paul.
While some of the changes are driving me bonkers and discouraging me from writing, I understand the goal. I just wish it were easier for human quality raters to guess what GOOGLE's quality rating system wants. Except, as of course you know, we can't write to the algorithm; we gotta write to people, the readers who are actually interested in our work.
I like your second prong approach in fighting Panda, getting your message front and center and throwing down the gauntlet to Google to say, "look, we've got good content that you're ignoring, and we're trying to weed out dross manually -- YOUR TURN to find the good content instead of killing it." Unlike us little guys, you actually have a small hope of getting that complaint taken seriously by Google's algorithm designers.
It's still hard.
And that comment about better quality content earning less than thin content is illuminating.
You make an interesting point about whether MTurk raters should rate based on potential traffic vs universally accepted quality standards (you know, good writing, actual facts, etc.). If all websites are eventually designed to attract traffic and drive revenue rather than provide quality information that is readable, we will have yet another dumbing-down of language standards. Sad.
I'd rather sacrifice some traffic in exchange for maintaining good quality. I cringe to think of where we are headed - I already get students (in my university courses) who write as though they're texting on their phones and who have trouble creating a thesis statement and presenting their argument logically.
I'm inclined to think that comment about thin content is untrue. And if it is true why are long articles being pushed so hard in the new hubtool?
Will - I am similarly confused about the emphasis HP places on producing content that has more text, more capsules and the other features of what HP refers to as "Stellar Hubs," if, as was stated in the interview, the revenue is lower. It's sad to think that higher-quality content draws in less revenue. But it would be even more disheartening for writers on the site to focus only on what brings in the dollars rather than on producing excellent work.
@Paul E - Great job on the interview; it's tough to be candid about something as personal as the health and viability of your brainchild (the site). Your answers were interesting and helpful, and it's good to hear that HP is committed to moving forward and finding what will work and succeed in the current SEO environment.
I may be way off base, but it seems to me that 1,000 visitors to short crappy hubs will pay more than 1,000 visitors to high quality hubs.
The kicker is that you can't get visitors to short, crappy hubs. The total number of visits to a better hub more than makes up for the lower income per visit.
At least that's what I get from the comments, notwithstanding that eHow often ranks better than HP.
Quality of the writing has nothing at all to do with the potential income, the one and only way the two are related is by traffic. 1000 views of crappy content with a high paying keyword will pay more than 1000 views of good content of a low paying keyword. Quality will effect number of views, nothing more.
In just one way I would disagree. Quality content provides answers; poor content encourages visitors to click an ad for what they're looking for. I've actually seen suggestions that we should write in such a manner as to encourage the reader to click through to an ad or another site for a full answer.
In that way crappy content can actually earn more per visitor, which is what I understood Paul to be saying.
There is a statement from Paul E. somewhere here, might be on a hub he wrote, or might be on a forum thread, where he says that they've seen that lower quality pages have a higher click through rate than high quality pages. So I agree with wilderness, I think this is what Paul was referring to.
One entirely valid point that Paul made that has not been highlighted here is that Panda discourages content creation.
If you have an account that is in good shape, and has never been hit by Panda why risk adding more pages?
If you get hit by Panda in the early stages of learning to write (because your pages are not great) that will seriously discourage you.
If you happen to have a half-decent brain, why invest time in a career which is so precarious?
I have to say I agree with you on this, Will. I have no impulse to write anything more to make money on HP. It's just for enjoying writing my fiction now. Too frustrating for anything else. Let the suckahs worry about the hoops!
Tell me about it. I was just 8 months old when Panda kissed HP, and looking for my third adsense payout. Instead of taking only 2 months between payouts that one took closer to 4. Discouraging wasn't the word.
I find this article very discouraging. I've only just started here and it just feels like there's no use in continuing.
It is a lot more difficult now than it was pre-Panda. At the same time, if you get it right, you can make good money.
And of course, Panda affects everyone writing online, not just HP. I was not criticizing this site.
Personally, I do OK here. I'm perfectly happy with traffic. Income from my accounts is very acceptable.
Ditto what Will says. It can seem like a roller coaster at times. But still worth it in my experience, whether you are here for fun or profit, or both.
It's good to know HP can still be a rewarding place to write (fulfilling, and even financially rewarding).
In an effort to try and pinpoint my own lack of motivation lately, the things that stand out are the hours I have spent editing hubs and trying to preempt the idling process (trying to guess which hub might be idled next, due to whatever the site's computer thinks is 'low traffic'), and an intense several weeks of filing copyright infringement complaints (mostly due to Mr. Dan Gordon's huge back-linking enterprise, which hit concurrently with the idling thing. These two chores have absorbed a ton of my time - and while HP is not the source of the copied hubs, the added pain of editing hubs that have already been vetted before (including through the AP process) has been more than frustrating.
At first, I was in favor of the idling process - especially for new hubs and new writers, or for junk that had lingered on the site for ages. But when experienced writers with proven skills are put through the wringer, it is not serving the site well. When hubs have already been vetted and then go idle because they haven't had a chance to build up traffic, it's not helping the site. And it's driving off the writers. I can either spend time constantly editing hubs to keep them fresh and filing complaints about the Dan Gordon types, or I can spend time writing new (and good) content.
The site absolutely needs a way to rid the inventory of bad hubs and bad Hubbers. I still support the idea of a hub going through review before it's published. I also support (even more than appears to be implemented here) a review process of all new accounts. But we probably have hundreds of people signing up every day, so that's a real challenge. However, when writers with a known and good track record are put through these frustrations, it will eventually drive away the people you want to have here. It's already done that, and it also drives away the content you want to retain.
In looking at the interview (which was great), it looks like HP still has a good SHARE of the activity on Google. While the site hasn't recovered completely, it's taken less of a longterm hit, and the sheer numbers are much healthier than some of the other sites mentioned or listed in the graphs. This is good - very good.
The community here, of staff as well as writers, is great. There are certainly some 'interesting' forum battles, and there are many outspoken Hubbers, but that's to be expected. Good writers are able to voice themselves forcefully - you would not want a bunch of 'yes people,' because the quality of the site would suffer.
I just hope the needless things that can chip away at our motivation can be tempered a bit. There are a ton of good writers who deserve a 'free pass' on the idling thing for new work, and who also deserve not to be subjected to having their writing idled before it has even gained traction.
Marcy, I think you put into words what many people seem to be feeling and expressing every day in one clear post. As a relative newbie, I'm observing and learning and yours seems to be the collective feeling from both old and new hubbers.
Even as a new hp writer, knowing that just passing the QAP is not enough and further not knowing how long I have to see if my hubs can gain traction and a place in the system leaves me worried.
Hopefully there is something in the works with clear guidelines of a timeline for idling and a warning system with at least a few weeks' notice. (If the idling is to stay in place.)
Yes, lower quality content has a higher click through rate on ads than high quality content, however I would only put the effort into high quality content.
@Will Apse I think there are multiple types of risk, but I wouldn't stop creating or updating content out of fear of Google. I think we are really close to breaking through.
Now I'm really wondering what the definition of "high quality" content is. Does it mean what it means out in the real world of writing, or what it means here and on Google?
The last paragraph of your interview was an interesting insight:
"This economic change may create an opportunity for genuine, independent enthusiasts (who enjoy a lower cost structure and the ability to vertically integrate all aspects of the content creation process) to succeed."
This seems to indicate that you believe that 'capable' people should have their own Websites?
Would you care to expand on your statement ?
In the case of Izzy, CMHypno, and myself, our traffic fell 85% literally overnight, right after we switched to subs. So yes, the switch did indeed make a huge difference in traffic.
Question (for Paul, Simone or anyone else who might know): Is there any outside chance that Google cares at all about traffic that comes from the 'you read my hub & I'll read yours' clicks that happen when large groups of writers trade links on FB, just so they can get clicks? (Not ad clicks - 'views' on the page.)
To me, this is artificially generated traffic - and we have a lot of it here. It's one thing to promote your work via Twitter, or whatever, but those groups exist primarily to increase traffic for each other. I'm just wondering if Google analyzes traffic that granularly, or if they care. Or if HP sees that as a concern.
One point about the idling - those of us who do share via Twitter, Google+, whatever, often wait until a new hub has been indexed - this cuts down on the overall traffic the hub will get - it makes them less timely (if that's a factor), and the writer can easily forget to go back and do that step. And, it's harder to link internally, to our other hubs.
I was also interested in some information I saw on the chart in one of this article's companion articles. It was a chart from data compiled by Searchmetrics on 22 businesses that had lost traffic and visibility with Panda. Each company was given an "SEO visibility score" to show how visible that company's articles are or were in search results, based on some specific keywords that are tracked.
Of the 22, a little handful (5-6) including Hubpages lost only around 50% visibility with Panda. The others lost much more.
Two of the twenty-two have improved their visibility ranking compared with the month after Panda (but not compared with the week pre-Panda).
Of the other twenty, Hubpages comes closest to being back where it was in March of 2011, a few weeks after Panda, with only a 5% decline in SEO visibility since that time, where the other 19 have 27-100% loss in visibility since then.
Obviously, visibility and traffic are not the same thing, but visibility in search results increases the possibility and the likelihood of traffic. Also, the metrics are based on specific keywords, and that is bound to play some part in overall visibility.
So, I don't want to make more of the chart than it is worth. But I'm still encouraged by the fact that HP has improved to the extent that it has. Keep it up, y'all!
I caught that, too, and it is encouraging. Add in that actual traffic, as opposed to visibility, has increased rather than decreased that 5% even while dumping thousands upon thousands of poor quality hubs and there is definitely hope.
See I read that in a very different way. Yes it is excellent that HubPages decided to "fight Panda" and keep more or less the same business model of publishing user generated content, whereas most of the other sites basically gave up, and either faded into oblivion or completely changed what they do.
However, I find it quite scary that HP SEO's visibility is 5% lower than it was in the weeks after Panda. Panda was pretty disastrous to HP with a huge fall of traffic. The fact that things are 5% worse now than in those dark days, is far from encouraging.
But what I really care about is traffic to my subdomain. This has increased nicely this month, which is good. As you point out SEO visibility is not the same as traffic, so I am hopeful that things will get even better.
Incidentally I don't really think that it is very realistic to go back to the pre-Panda days. It seems to me that Google was much less sophisticated then, with far more 'mechanistic' criteria. So writing on a big, aged domain, with a high PR and a million pages indexed gave hubbers an SEO advantage that they didn't really deserve, or at least some of them didn't. At the same time, hopefully, we are now at a disadvantage that we don't deserve (a penalty), so if that could be removed, we would do better, although probably not as well as pre-Panda.
Go back to pre-Panda? No. Get back to that traffic? Yes.
HP has dumped tens of thousands of hubs in the last few months, with no real decrease in traffic. My own traffic is some 5X pre-Panda, with only 3X the number of hubs, and 60 of those are still less than a year old.
Just for clarification, aa lite, nowhere did I or the article say that "things are 5% worse now than in those dark days." The 5% figure referred specifically and only to SEO visibility based on specific metrics.
If you go to http://www.quantcast.com/hubpages.com you can see the chart of Hubpages traffic. At the bottom right of the chart, click on "Custom" and use January 2011 as the starting date. You will see the huge drop from ~800,000 pre-Panda to ~400,000 (and later, even down to ~300,000) immediately post-Panda and shortly thereafter. We have had frequent fluctuations since then, including one brief peak that looks almost like pre-Panda levels. So far, in the past few months, traffic has not been as high as in late 2011-early 2012. But it is absolutely much improved over the immediate aftermath of Panda 1, and we seem currently to be hovering somewhere in the 500,000-600,000 range.
I agree with all that. But a lot of your post was about SEO visibility and the fact that it is only 5% worse than after Panda, and then you said "But I'm still encouraged by the fact that HP has improved to the extent that it has.". Sorry if I misunderstood.
by Dale Hyde 6 years ago
Right at one third of my hubs have been idled over recent months here on HubPages. As I look at the numbers and reflect upon the time spent creating this unique material, I am sadly disappointed. It is unpleasant to look into the future with so many people having this same problem and...
by Cindy Lawson 6 years ago
How is it that a hub on the 'common mistakes new hubbers make' can suddenly become not featured, in spite of the fact it has had 9 views in a day, 24 views in 7 days and 35 views in the last 30 days? I only replied to comments on it both 42 hours ago and 22 hours ago (from two different hubbers)....
by Yvonne Spence 6 years ago
I love HubPages, I feel that I have learned so much here and my confidence in writing non-fiction took such a huge leap when I was invited onto the Apprenticeship program last year. I have met some wonderful people here - both staff and other Hubbers. But as may other people have said, the idling...
by Steve Andrews 6 years ago
I keep on getting hubs idled despite tweaking them, changing titles and adding better keywords. I have got used to moving them to other sites such as Wizzley, Xobba and InfoBarrel but it seems to be an ongoing problem. If I keep on moving my idled hubs eventually I will hardly have any left!
by Janis Leslie Evans 6 years ago
I've been following the thread re: problems with the unfairness of idling hubs for low traffic. I went to FAQ and LC to find what criteria determines one getting idled. I couldn't find anything. I know it's obviously about low traffic but what is the measure? 1. Low traffic or no traffic for 30...
by Will Apse 6 years ago
There is a lot of SEO related stuff about Panda in these forums, so here is something about quality and the kinds of content Google is trying to find and offer to searchers:It comes from Amit Singhal, Google Fellow and High Priest of search.Would you trust the information presented in this...
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.|