It's no secret that many Hubbers have been hit by the recent Google 'Panda 4.1' update. Panda is one of Google's key quality algorithms that assesses your content for certain signals that indicate that your readers are happy with the site and user experience offered. If you noticed a sharp traffic loss after the 25th of September, Panda is the cause of your traffic loss.
There's no clear consensus as to what you should and should not be doing to recover from Panda, but there is definite correlation between a site with a lower quality rating and certain site metrics. The most important of these metrics is bounce rate, which is a percentage telling you how many people left your site without reading the content.
Bounce rate has a direct correlation with Panda. Sites that have a higher bounce rate are more likely to suffer from a traffic loss due to Panda. This is because Google sees bounce rate as an indication that searchers have not found what they want or have been met with a poor user experience that has turned them away. Google decreases the quality rating for a site with a high bounce rate, causing it to lose search ranking and traffic.
How to calculate your bounce rate
If you've been hit by Panda, or you're worried you might be hit in the future, you should be looking at your bounce rate. To do this, you can use Google Analytics. From the side panel, open the acquisition tab and access the channels screen. You can then proceed to 'organic search'. This will tell you the calculated bounce rate for all views derived from search engines during the past month. If you've been hit by Panda, you'll likely see a bounce rate of 80% or more.
This isn't good because it shows that the majority of your traffic has returned to the search engine immediately after arriving at your site, or in this case, your Hubpages subdomain. This is a very good signal to Panda that search users aren't happy with what they have been served by Google, and this is why Google will use it to penalise your site.
Why is your bounce rate high?
There are many reasons why your bounce rate can be high. The primary reason is that searchers aren't finding what they want, immediately, once they reach the page. If a searcher thought they were getting an article about 'cleaning an oven', but they can't see any cleaning information because you've started out describing how an oven works, they may click back straight away.
This is why Hubpages recommends you keep a hub relevant to its title and answer a reader's question straight away. You really want to stress that your reader is going to discover what they want to know, right from the first paragraph. You can add in extra related information, but try to introduce it later on and not let it detract from the primary focus.
Keywords and bounce rate
Due to the way Google sends you traffic, using a variety of ranking factors like keywords and backlinks, searchers may stumble upon your article looking for something entirely different to what you're offering them. If you are writing an article about 'how to set up a new computer', you would want your primary keywords to be related to that title. Preferentially, you'd want the main keyword used to be something like 'set up a new computer'. You want to avoid repeating phrases like 'choosing an OS' because your article isn't specifically about that topic.
If you're inadvertently repeating unrelated phrases or words, and these are ending up comprising a more dominant density compared to your target keyword, you'll find searchers end up landing on your article and they're looking for something entirely different to what you're providing. They return to Google straight away, and Google receives data that tells it that your site isn't liked by searchers. Keep your content relevant to your topic, but keep your keywords just as relevant, because this is how Google sends traffic to your site. You want targeted readers that will find what they're looking for instead of leaving straight away.
More causes of a high bounce rate
There are so many other causes of a high bounce rate, including spelling and grammar, site appearance, amount and type of ads, how knowledgeable and professional the author is on the subject, etc.
Whilst factors like spelling and grammar aren't likely measured directly by Google, it's fairly obvious that a reader is going to return to search results fairly quickly if your article contains many mistakes. This is why you should be checking your spelling and grammar and trying to maintain a high standard of English when you write. If your readers cringe when they try to read your article, they will bounce.
This also relates to site appearance, pictures, and advertising. If your article is a wall of text, a reader may look at it and think "No...that's too hard to read". Try to break up your text with pictures and other elements. Pictures are particularly beneficial because pretty pictures will draw the attention of the reader and keep them on the page for longer. If the picture helps explain the content, or relates to the content in a way that directly enriches it, this is even better.
Advertising plays into this because many ads just look spammy and encourage a reader to leave the site quickly. If a reader finds a certain ad offensive, they may even leave because of that. The site appearance as a whole needs to be attractive in order to keep readers from immediately leaving.
Assessing your landing pages
The pages your search traffic arrives on are the key factor in your bounce rate. If you have certain articles that receive the bulk of traffic, you will want to focus on improving the user experience for these pages, primarily. I believe this is what Hubpages was intending on achieving with its HubPro program.
The first thing you should look at though, is not the user experience, but the keyword breakdown. Using Webmaster Tools or Analytics will show you which keywords are driving the bulk of traffic to your site, and you will be able to link this back to your landing pages. If the majority of these keywords don't relate to your content at all, you need to focus on changing the wording of your article to better explain to search engines just what exactly your article is about. This will help stop searchers ending up on your article believing it is about something else. These searchers immediately bounce, and you want to keep your searchers on target to limit this.
After you've done this, that's when you should proceed to improving the user experience. Hubpages provides plenty of information about how to do this, and the whole 'stellar hub' target helps you achieve this. You can also find a wealth of information about user experience all over the internet.
Evaluating landing pages with Google Analytics
To see your landing pages using Google Analytics, access the behaviour tab, go to 'site content', then to 'landing pages'. This will provide you with a breakdown of the top pages for the past month (Or whatever period of time you have set). You might also like to set the secondary dimension as 'traffic type' to further purify the results into landing pages based on organic search, direct traffic, referred, etc.
Evaluating engagement with Google Analytics
Another set of stats you may want to look at in more depth is your view times, otherwise known as engagement. To do this, access the audience tab, then 'behaviour', and click on 'engagement'. This will breakdown your traffic into time periods. The first time period of 0 - 10 seconds is your bounce rate.
You can also 'add a segment' to further divide these statistics down. Add organic search in and you can look at engagement data for organic search specifically.
What can you do with this data?
Your search engine landing pages are the primary factor in bounce rate, because they receive the bulk of your search traffic. Set them up in a way that guarantees that the reader will stay on the page and Panda will leave you alone. Locate your landing pages and assess what keywords searchers are using to find it, as well as the on-page user experience and readability.
Engagement, on the other hand, can be used to track your bounce rate improvements. This will show you whether your efforts are having any effect and help you focus your time and energy where it's needed most. You want to limit the amount of people that leave in the 0 - 10 second time period, and you want to push more readers into the lengthier time periods to really stress to Google that searchers love your content.
All of this goes in line with what Hubpages has expressed time and time again, as well as what they attempt to do with their stellar hub ideal and HubPro program. The only difference is, having the actual data helps you improve your subdomain without clutching at straws. You'll be able to see whether something you're doing is working. Hubpages doesn't recommend using plenty of brilliant images, long copy, and other quality elements specifically because Panda is analysing that, but rather because elements like that help keep readers on your page for longer, reducing bounce rate, which is what Panda does definitely analyse.
Incidentally, I've noticed that when comparing my Hubpages subdomain to my blogs and websites, although there may be content in the same niche or a similar niche, written in a similar way, organic traffic bounce rate is roughly half that of Hubpages. I can't specifically say why this is the case, but readers simply bounce more often from an article on Hubpages than one that is on other sites. This is likely due to site appearance and such, but I finish drawing conclusions there.
Thanks Maffew for this wealth of information! I've learned that people stay on my HubPages account for approximately two minutes, however, my traffic has diminished after this last Panda update. I now need to focus on the hubs with no traffic and fix their appearance with keywords and more pictures or a more relevant video.
I do get your point and thanks for the info. However, kindly visit this link to better understand what bounce rate is;
https://support.google.com/analytics/an … 5491?hl=en
Wow, thank you for this! You should have posted it as a hub
First of all, thanks for this information, it is very valuable and you have obviously been doing your homework for quite some time.
But I have a few issues with using Bounce Rate as a measure.
My understanding is that Google only measures Bounce Rate once someone has moved on to another page on your site. Meaning, they might have been on your article for ten minutes, got what they wanted and were happy about it, but then just left the page. This means user experience is good, but measured as bad. This is definitely the case because average time on page for many of my visitors is high but my bounce rate is high too. Now, some time ago it was said by admin that time on page is more important than bounce rate. Also, many people on HP don't write on single subjects on their subdomains, we tend to bounce around in subject matter; this inevitably leads to a high bounce rate because readers get what they need and leave, they have nowhere else to go on your site; of course, unless you have a subdomain on a single subject, link Hubs or use the Hub Groups feature.
Of course, I know what all this means. Subdomains would have to concentrate on one subject. Which, actually, is what I plan to do from now on; have an account per subject matter. Plus concentrate on my own sites which are, of course, each concentrated on specific subject matter.
I do agree with you about the session time. I think Google uses both to some degree. We'd probably see the most benefit by trying to get people to stay on our subdomains for as long as achievable, as well as view more than one page wherever possible. Compared to sites where I've wrote on a variety of subjects, my single subject sites have always seemed to perform better and attract more organic traffic.
I have often asked about bounce rates because mine have never been good, and lately have been worse than ever. What I have been told by other gurus here is that bounce rate can mean people are getting exactly what they want from your article very quickly due to skimming, etc., and can quickly move on.
I do not understand what I am doing wrong as my articles are very relevant to their topics, have photos, videos, etc, etc and yet my bounce rate, according to you, is killing me off. I am totally confused about all of this and wonder if you might take a peek at one or two of my articles and tell me what you think.
How high is a high bounce rate? Around how much percent is considered high?
The average bounce rate for a content site is somewhere around the 40 - 60% mark, so that's where I'd suggest aiming for. I'd imagine Google would assess a website against similar sites, so it may be best to aim for whatever is usual in your niche.
That's not what I'm suggesting. The pages that attract most of your traffic are your landing pages. If searchers click through to these pages through search engines, then quickly return to the search engine without viewing subsequent pages, that's a bounce. If they don't spend any great amount of time reading the material, there's a lack of engagement and interest. Google can see this.
The reason I believe you should focus on your landing pages is simply because they receive the bulk of your initial search engine traffic and will have a stronger effect on your site as a whole. If you change these pages in a way that encourages readers to read more of your articles, and remain on the site for longer, it will have a dramatic effect on engagement metrics. The main thing is that you prevent those readers from clicking back to Google.
I do understand the difference between a bounce and an exit. A bounce is when a user leaves from the first page they visit on a site. The exit rate simply refers to the percentage of people that are leaving your site from that page. A page can have a high exit rate but a low bounce rate.
I did mash session time in with what I originally said, because I believe that's important to Google, along with bounce rate. Maybe that's the source of confusion because I didn't really separate these metrics out. Because some queries may be quick and easy to address and it's not uncommon for users to get their answer from the one page and then want to leave, Google would be more concerned with view time here. A quick departure would suggest an answer wasn't found, whilst a long stay would better suggest quality content, even if the user eventually bounced back to Google from the article.
Thanks for explaining extensively, Now I know. It seems though that my bounce rate is way above the average. I have to work on that. Thanks again.
It also seems that the more backlinks a site has, the more aggressively it is judged by Panda. That's rather interesting if true.
I would expect the bounce rate to be higher on HubPages because the readers would go to another hub from the related hubs links at the bottom of the page. What would you consider to be a normal or optimal bounce rate on this site?
Panda only affects web sites. not individual pages within a website.
I agree. If you got hit by panda your whole website goes down.
I wasn't suggesting that traffic to individual pages will decrease because of Panda. I'm saying that your bounce rate and session times of your pages are correlated with a Panda penalty and probably used by the algorithm as quality signals. If your bounce rate for organic traffic is high, and your average session times are low, this can be used by Panda as a way of deciding whether content is low quality.
When I talk about landing pages, I'm talking about the pages that attract the bulk of your traffic. If you can decrease your bounce rate on these pages, this has the biggest effect on your site as a whole.
Great advice and information, thank you. Food for thought indeed. I wonder, some hubbers may have many hubs, some with high bounce rate, others with decent rates. My question is : Would it be wise to open another account, move the hubs that are doing better to the new account and leave the poorer performers on the original?
Thanks, Maffew. You are one the best at giving relevant information that makes sense, is concise, and easy to understand and apply. But I'm tired. I think I need a break. My numbers trying to improve with stops and starts, but today's are dismal. Time for me to inhale, exhale, and come back, maybe next week or next month. See you guys later. Good luck and be blessed.
I wish you well janshares. Please come back as soon as possible with your positive vibes and understanding. And your poetic soul!
Don't give up hope Jan. If you've had success in the past, it shows your content is good. If all else fails, you can always look at starting your own website and transferring it there. You have more control that way and don't have to worry about Google's vendetta against Hubpages, or other authors potentially dragging you down. I've certainly been considering doing this in the past few months.
Well using Analytics and comparing October with September, it looks like page load times are about 5 - 10 times slower after the server transfer. There must have been some sort of server issue on October 20th, because page load times on my hubs were about 90 seconds on that day.
Using Pagespeed Insights, what shows up for photos is really just a suggestion to compress image files to reduce their file size. The photos specified are mostly profile photos and other site images. You may see some of your actual hub images there, but it doesn't look like anything serious.
It also shows that there are scripts that load first before the actual page content can load, and suggests using asynchronous code, which would mean it then loads at the same time as the rest of your site. Hubpages probably has those scripts set to load first for a reason. It looks like they're mostly involved in statistics and advertising.
The leverage browser cache suggestion means Google is suggesting you set up an expiry time for certain resources, like pictures, meaning if someone was to visit your articles more than once in a set time period, the page would load quicker for them because it would already have those elements stored in the browser cache instead of having to load them again. This is another issue Hubpages would need to deal with and probably have already looked at and decided against for whatever reason.
I'm not sure what the optimise CSS delivery suggestion entails. It links to CSS style sheets that Hubpages uses to set up the appearance of the site, but my knowledge of CSS is lacking.
Today's HP numbers are dismal. However, I think there may be a problem with the HP counter. In GA my views are off by 100 views in HP they are off by 400 views. There is always a discrepancy between GA and HP views, but typically not this great of a gap. It seems something is out of whack today.
I did notice that too actually. I'd been paying closer attention to my Analytics because of the stat problems over the past week, and my Analytics page views came out fairly consistent with the rest of the week whilst today's stats on Hubpages seemed strangely low.
Webmaster Tools tests page load time. In testing two hubs, they found problems with photos and some other technical issues on mine. Maffew, could you test there, and give your opinions. Sure would appreciate the effort.
Thanks for great info Maffew. You give precise instructions.
Thank you Maffew for a well written and easy to understand informative piece. I find that my bounce rate on a couple of my websites are between 30-40% but on HubPages, its in the 80's.
If you review Google Analytics you will see you have individual bounce rates for each and every Hub.
They vary hub to hub.
My bounce rate is around 91, but I receive almost 300 visitors per day having 38 hubs. All my traffic comes from google i.e., I have organic traffic. What do you think about it, James?
Good article. Should cut paste and make a hub out of it. Also hubpages is non-responsive so mobile and tablet readers bounce off. As for looking at keywords these are for the most part not represented anymore.
It doesn't really fit in with my niche and I really just wanted to share something which could be helpful in Panda recovery, so a hub is out of the question.
What do you mean about the keywords though?
Yeah, I just found out about the non-responsive thing testing one of my hubs on the emulator at http://quirktools.com/screenfly/. I assumed Hubpages had gone responsive long ago, and was quite shocked to find out they hadn't. No wonder the PageSpeed Insights scores are horrible for mobile. In Google's eyes, we're not optimized for the devices that make half of our visits. (Half of mine, anyway.)
I don't see a huge difference between Bounce Rates for Desktop vs. Mobile. Time Spent seems to be higher for Mobile for my most visited hubs. Varies on others, but overall, it isn't a red flag on my particular subdomain. However, I realize we have to interpret cautiously and take both stats with a grain of salt, given the site structure and the fact that most people don't visit a second page.
This mobile thing, though, might be an issue. I'm no expert on Mobile App/Mobile Site vs Responsive design, but I would think it'd be best to comply with the one Google says they prefer, which is Responsive design: https://developers.google.com/webmaster … es/details
Just my two cents. I obviously have no insight into the business decision to go the App route.
If Google prefers a responsive display, it's probably a good idea to follow that. I wouldn't know whether that is impacting views to the mobile version of Hubpages. I had actually believed Hubpages had a responsive display too, but I don't tend to use my mobile on Hubpages. I prefer to access the desktop version of sites, even on mobile, and this is something that's made it harder for me to optimise my blogs and websites for mobile, because I think mobile sites are often clunky and hard to navigate.
I had a quick look. Are you using keywords in your copy? Going off the two articles I looked at, the most dense keywords were somewhat random, like "items you" topped the list of 2 word keywords for one article, whilst "of an rv" was on top for 3 word keywords. This would suggest your traffic from Google isn't as targeted as it could be.
If you have a look at your Webmaster Tools queries, that should give you a good idea of what terms people are using when they find your articles. If the terms they are using aren't represented by content in the article, they may leave immediately or fairly quickly because they can't find what Google said they were going to get.
You also might be able to focus more on internal linking. Take your best, catchiest, related articles and link to them from within your landing pages. Ie, if you have a page that attracts a lot of traffic from Google, try to get readers of that page to click through to another interesting article. This will increase time on your subdomain, page views (And thus income), and decrease your bounce rate.
I'd also suggest looking at your engagement data in analytics. Specifically, the amount of visitors that view your content for less than 10 seconds. Having a high percentage of people leaving after 10 seconds is a sure fire way to make Google think searchers aren't getting what you want. If you can't decrease your bounce rate, you should at least try to increase the time spent on site. A high bounce rate will draw Panda, but long views would show that people are in fact getting what they want, even though they may leave after that one page.
Thank you for taking the time to look at my articles. Since I am now writing strictly about recreational vehicles, I have been warned to be careful about using RV too much, so I use as many synonyms apossible. I use terms like "items like" as a way to limit my keyword density and avoid problems.
When I check WMT, the term RV is always in the search query, and when I check engagement, most of the time, especially on certain specific topics, engagement is 5 stars. I also interlink as much as possible wherever it makes sense...most of the time it is two links, but sometimes as many as six.
Where are you finding the most linked to terms info? On WMT? I never thought to look at two word/three word keywords...I just always went to adwords, looked for terms that were searched most often, and tried to use them. I am not good with keywords, so perhaps that is where I am failing.
Also, I was doing extremely well after the May update until August 26 when my views took a sudden dive, so I tend to think something happened there that made me lose my rankings...but do not know what it was. So my problems are not the result of Panda, Penguin, etc...as far as I can tell.
That's what I've noticed on my websites too. Millionaire Tips pointed out earlier that users clicking on other people's hubs might influence this.
Yes, they do. That's why I keep reinforcing the point that any effort taken to decrease bounce rate to the hubs you have that attract the most traffic, will have the most significant effect on your site-wide bounce rate. The site-wide bounce rate is the average of the bounce rate to all pages. Your pages that attract the most traffic will have the biggest influence on bounce rate as a whole and that's where you should focus on.
I couldn't say for sure. If the high bounce rate is the result of readers not finding what they want, it may be better just to work on those hubs and improve their rate rather than ship them off. Transferring them to a different account would result in a better site-wide bounce rate for the first subdomain though, because the average drops down, but you'd lose whatever ranking you had gained, and any backlinks to the hubs would no longer be valid.
It'd probably be better just to work on improving the user experience and relevancy in the long term, unless they don't seem to get many views anyway. But even in that case, if they're not getting many views, they're not playing a bit part in your site-wide metrics.
I'm not too sure Rafiq. Without knowing what keywords you were targeting, who your competitors are, or what niche you're involved in, I can draw very few conclusions about what your traffic could potentially be. What I can say is that prior to the recent Panda update, my most popular hub had consistent traffic of between 800 to 1000 views per day, and this dropped down to about 300 per day after the update. If I'd never known how much potential it had, I'd think that was pretty good for one hub, but I know it could be receiving so many more views because of what I had seen in the months prior to Panda.
Based on that, for all you know, your 300 visitors could potentially be 3000 if they were ranking at their maximum potential. I don't know whether your hubs are ranking at their maximum potential, or whether you could be getting much more out of them. You could probably work out whether this is the case by looking at any past data you have, as well as your average position in search.
Great info Maffew. So good of you to take the time to share it!
The only thing I would say, is that we have to a bit careful looking at a high bounce rate as negative on a site like Hubpages.
There are loads of ways a visitor can bounce.
1. They can go back to the search results. This can be the worst way to bounce because Google will track how long the searcher was on our page, and if they hit back very quickly they may decide our page is not fulfilling the expectations of the search query and consequently, reduce our page's ranking. But, it does depend on the search query. Some queries only require a short answer.
The visitor could also read the whole hub, feel completely satisfied and then return to the search results. This would also be a bounce.
2. They can click through to another subdomain or an EC hub which is on Hubpages main domain. These are counted as other sites, so these are bounces too.
3. They can click on an ad, whether it be adsense, ebay, amazon or another affiliate link if you use them. This is the kind of bounce we want because we stand to earn from it (unless they're clicking on the related search ads which we get no income from).
4. They can click on a link to another site that we've linked to and leave Hubpages.
5. The visitor finds what they wanted on the page and then closes the browser or leaves it open for 60 minutes while doing other things.
My own bounce rate on HP is around 90% and it's never been a problem.
That's not to say it's not worth trying to keep people on your sub through careful interlinking - it is, I just wouldn't worry about it overly.
by Nathan Bernardo4 years ago
I have hubs within a couple of niches which I don't cover all that much; meaning, I don't have that many hubs in those niches. A good number of those hubs have a high bounce rate. So, I'm considering unpublishing those,...
by Will Apse5 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...
by Wendy Hughes2 years ago
I know traffic is a problem everywhere it seems; I'm not talking about low traffic, here. We all need to be aware of the HP's photo capsule may be affecting our bounce rates. There is a real problem if you use the photo...
by Susan Britton4 years ago
I have a high bounce rate in the eighties and I have read it should be at 50 . Is this true? Can anyone share how they get their bounce rate to the fifties?
by Steve Andrews4 years ago
When it was first introduced I was annoyed by it but made an effort to tweak my hubs to get them out of Idle status. Now, a whole load have got zzs against them again and many of them are hubs that at one point were...
by KiaKitori6 years ago
I jus finished rading this article and thought to share it with you : http://www.seo-theory.com/2011/05/06/ch … algorithm/It say that this Panda technology is the foundation of the new internet search, and it is...
Copyright © 2017 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.