Websites Affected By Zombie Traffic - Observations & Analysis

A fresh take on several existing threads is welcome. Is there is list of specifications/parameters which define zombie traffic? Having a definition or expressed theory will make the discussion more concise.

I too am curious reseller. Exactly what is zombie traffic? And from your experience, how does it affect a website?

I would define the so called Zombie Traffic in general as: Non-productive Google organic traffic.
At present there isn't an agreement within our WebmasterWorld community about the origin of the said traffic. However most webmasters whose websites have be affected by Zombie Traffic have rejected mis-mached search queries as possible origin of the said traffic.

It seems that websites affected by Zombie Traffic illustrate mostly the following characters:

1 - On/Off conversions pattern.
2 - On/Of ranking of main keywords.
3 - High ranking of main keywords on Google SERPs doesn't necessary mean more productive organic traffic to the affected websites.

A problem with patterns is that humans look for them in an attempt to explain the world about them; in doing so they are often seen where they don't actually exist, i.e. Are an artifact of something(s) else. I am NOT saying that is the case here just that it needs always to be kept in mind when doing analysis.

Also that we as a group don't have sufficient data because webdevs, myself included, don't like to share too readily. However, what would be extremely useful to know are things such as:
* typical daily site Google traffic within certain bounds, i.e. less than 100, less than 1000, less than 10000...

* conversion traffic IP geo-regions, i.e. countries or at least continents.

* apparent zombie traffic IP geo-regions, i.e. countries or at least continents.

* time of day, day of week, of apparent shift in traffic.

* historical conversion rate of Google traffic.

* site revenue percentage breakout between eCom, ads, aff.

Without the above and similar data sharing I don't see this thread becoming anything different than previous threads.

Without actual data we are replicating the blind men and the elephant .

...most webmasters whose websites have be affected by Zombie Traffic...

What is the affect on a website experiencing zombie traffic?

What is the affect on a website experiencing zombie traffic?

Websites which are subject to non-productive Zombie Traffic lose conversions and hence lose revenue.

• So the problem is that these bots do not convert?
  
  
• Or are these real people you are concerned about?

If they're real people check if it's Google Images traffic.

Or if they're actually hitting your site from regular search check them against heat map software. What are their patterns?

• Are they researching and declining to go further because they're put off by prices, slow loading, shipping costs?
  
• Does the site make it difficult to compare products?
  
• Are your images not optimized?
  
• Does the site render like crap in mobile?

Are you ranking for irrelevant phrases like software updates you don't offer for products you offer?

If they're bots then they really don't matter, they're just probing for vulnerabilities or scraping. If it's Google Images traffic it's a neutral situation, doesn't matter.

If it's an irrelevant phrase thing then that's another neutral issue.

If it's about people looking for things you don't offer then maybe take steps to remove that or else upsell them.

Aside from that last one, people hitting the site for things you don't offer, the rest of the reasons seem trivial.

I'm just wondering whether Zombie Traffic has been affected by the two recent suggested Google Algorithm Updates of 24th January 2017 and 2nd February 2017.

Have the websites affected by Zombie Traffic been receiving less or more of the said traffic due to the two above mentioned updates?

@reseller am receiving more traffic but the zombies are still alive and active. Looks like the updates didn't get rid of the zombies.

@reseller I am getting less zombie traffic recently. My site is primarily a place for zombies. Overall, I now have less traffic.

For people wondering what zombies are, those are fairly easy to identify. The zombies come in on the landing pages and then disappear. They will not do any kind of interaction, not even accidentally. A further characteristic of zombies is that they come-in in clusters, almost as if google is trying to fill their zombie quota for a site. Once that's done, traffic disappears.

So if you are getting no traffic all day, then a bunch of them come in with short time gaps in between, you would have been graced by the zombies. Sometimes it's very obvious. For instance, zombies using windows Vista from wildly different part of the world would suddenly decide to come visit all at the same time.

@NoobOperator

They will not do any kind of interaction, not even accidentally.

How do you determine when an interaction is accidental and when it is not?

@NoobOperator

@reseller I am getting less zombie traffic recently. My site is primarily a place for zombies. Overall, I now have less traffic.

Thanks for feedback. Much appreciated.

@NickMNS, well live animals and humans are a bit random. They often do things you don't expect like clicking on things, have a jump up-and-down, whatever. Zombies are not like that, they are cold, dead, and mechanical. As I already indicated, they like to cluster around each other. So when the lemming train of the zombies come in, you know exactly who they are. Google's zombie train comes to my site every day, I can pick them out just like that.

Sometimes I can even induce the zombies to come in. When I am sick of waiting for the train, I just go to one of the web proxies and access my own site. Either immediately after, or within an hour the zombies roll in and do their cold, dead, flash by.

@NoobOperator
My question was very specific how can you differentiate between a real click and an accidental one?
I fail to see what randomness has to do with this question.

@NickMNS, my point being zombies don't click, accidentally or otherwise once they enter a site. As for deliberate or accidental clicks by real people after they enter a site, they are all good. I never had the need to find the distinction between these clicks.

I am guessing you want to say how I work out clicking on the search engine "search result" is a genuine click or a fake click ? I don't believe zombies come in by clicking, they are just programmatically generated web requests.

@NoobOperator

I never had the need to find the distinction between these clicks.

Then why make the distinction now?
A bounce is when a user visits a page and leave without any interactions.
If I understand correctly, what you are saying is that zombies are users that bounce, with the added distinction that this occurs in clusters.

Do you have any means of distinguishing between users that actually bounce and zombies?

One more question, if you have no means of determining whether a click is fake or not, how can you be sure that zombies are not also mixed in with what you believe to be actual users?

@NickMNS Do I have anyway to distinguish between the google zombies who come on their tu-tu's and an innocent passer-by human who happens to bounce at the same time ? No. But it's highly improbably humans will only chose to come in at the same time the tu-tu is arriving. Humans are more random than that and will come in at any time they please if they have a way in.

My site suffers from bad traffic days after I make some sales. Then there will only be tu-tu's arriving. After a few weeks of tu-tu'ing real people are let in and they interact and they buy. Then I get shut down again and left to play with zombies. Though google always ensure they download a couple of copies of my product before the shut down - this marks the beginning of the tu-tu traffic . It's a repeating pattern for a couple of years.

The following are the signals for beginning running of the tu-tu's (I have masked out some bits to keep privacy). I would be interested in anyone's experience of these google IP's:

access_log-20170206:66.102.8.28 06/Feb/2017:02:06:36 "/[censored]E.zip" "http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CDQ___AB&url=https%3A%2F%2Fwww.[censored].com%2F[censored]E.zip&ei=K9qXWNiR____0wPFZQ&usg=AFQjCNGGv94dK7zho1____sG3oTWsmM8QQ" "(Windows 10.0; x64) Chrome"

access_log-20170206:64.233.172.130 06/Feb/2017:02:06:42 "/[censored]E.zip" "http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CCwQFj____rl=https%3A%2F%2Fwww.[censored].com%2F[censored]E.zip&ei=MtqXWJm8M4SX6___sgE&usg=AFQjCNGGv94d____o161GDsG3oTWsmM8QQ" "(Windows 7; WOW64) Chrome"

[edited by: NoobOperator at 11:41 pm (utc) on Feb 8, 2017]

I got to ask ... how does g know users buy and they then squelch further purchases to manage a "quota"?

While ecommerce AS AN INDUSTRY is going gang busters in accumulated numbers (dependent upon local, national and global metrics based on general economics), the "average" "no brand" or "local eseller" is not seeing that same growth. Sites at <$1k a day are having the hardest time while larger sites just keep getting larger, sucking up all the "loose users" which once went "anywhere". Is this even a SEO question in the first place? Last time I looked there were no guarantees regarding traffic of any kind, much less CONVERTING traffic, and that applies to all search engines.

@tangor. I updated my above response with a little bit more info.

Google has adsence on my site. So they can track incoming paypal originated traffic. Without stating the obvious, they also have tracking of the buyers if they use chrome. Lastly, my personal belief is paypal either share their data or their systems are compromised. So the transactions are not truely private. Of course I have no proof and just my gut doing the talking. Damn clever gut don't you think ?

At the very beginning I had only amazon interested in my sales. Their response was a week. Google's response is in low hours.

@NoobOperator I checked my logs for the ip you listed. I do not have the exact IP's but I have several entries for addresses on the same C-block. All seem to be originating from India or Brazil and include (KHTML, like Gecko; googleweblight). Suggesting that these users are viewing the pages through Googleweblight, [googleweblight.info...]
Note that the IPs are as you stated Google ips, but the referrers are .co.in or .br.

This may explain the strange user behaviour, as the services strips out non-essential elements from the page before showing to the users. The question is, how does it determine what is or isn't essential.

Humans are more random than that

There is no such thing as more or less random. Either a process is or is not random. It is well studied and well documented that humans (in this case not the users but webmasters) are really bad at determining what is or isn't random.
See Apophenia here [en.wikipedia.org...] and Clustering illusion here [en.wikipedia.org...]

@NickMNS OK Mr Random Expert, using your random theory, can you explain why when I block those google IP's from downloading my product, a few days later chrome would pop up a "not commonly downloaded and could be dangerous message" for every downloader using chrome from anywhere in the world ? Where are these "indians" and "brazilians" getting their amazing powers from - they seem to have complete control of google ? I can imagine one possible answer is: they are script kiddie hack0rZ, right ?!

can you explain why when I block those google IP's from downloading my product, a few days later chrome would pop up a "not commonly downloaded and could be dangerous message"

It has nothing to do with randomness, is has to do with the fact that the users going to your site on the specific IP's you listed are accessing the site through some Google interface, such as Googleweblight, or Google Translate or any other type of service. (After posting looked deeper into my logs and found those ip also referring to translate). So:
A: the users are most likely not english speaking, or their mother tongue is not english
B: the Google service is messing your site layout up, so the user is being shown stuff that would typically be hidden or less prominent.
C: blocking resources signal to these Google services that what was originally shown, should not and so that content/button/link/download,etc... is removed and replaced with something even more outlandish.
combined these points together and the result is some really nonsensical usage patterns.

If you checkout the Googleweblight link in my post above, you can enter your url and see how it displays your site. Let me tell you, it makes a mess of mine. I use SVG extensively to display bar charts and other graphics and all that is gone. I also have a tab which shows the data in table format for those that can't see SVG. In Googleweblight both don't work.

@NickMNS you are not answering the question. What is the connection between my blocking google from downloading my product and google popping up warning messages for other people who want to download my product ?

Although I haven't look at your other posts. I am willing to bet Googleweblight appears in a lot of them. This is the go to answer for every complaint people have on google, yes ? Despite my user name, I am highly experienced and technical. If we continue down this path, I can make you look extremely embarrassing. Do you really want to go there ?

Despite my user name, I am highly experienced and technical. If we continue down this path, I can make you look extremely embarrassing. Do you really want to go there ?

Don't wast your breath or keystrokes in this case trying to explain something to someone who admittedly does not suffer from zombie traffic and has no access to the technical details we have. In the recent past NickMNS has suggested that bots are the cause of zombies, and if this is the case where did the real humans go? NickMNS suggested that throttling may be a possibility but we have John Mueller of Google stating no webpage traffic ceilings or traffic quotas are used (see [seroundtable.com...] ). If ones Google Analytics and WMC console data closely resembles overall reported Google traffic, its safe to say that bots are not zombies and that Google is not reporting spoofed hits as coming from their own domain in both areas of reporting. The whole bot theory was retired well over a year ago and can be easily ruled out by those that want to spend a little time in their traffic logs. Who knows, maybe he's trying to be helpful or just add some FUD to the conversation. Regardless, the blame still falls in Google's lap. With record profits, it's working out well for Google so don't expect a cure for the zombies or Google to ease up on their stranglehold they have over global commerce anytime soon.

What is the connection between my blocking google from downloading my product and google popping up warning messages for other people who want to download my product ?

Possibly a caution in that g can't see what's there so might suspect the worst and advise their users accordingly? I suppose there's a good reason for blocking g from the product, right?

@NoobOperator

What is the connection between my blocking google from downloading my product and google popping up warning messages for other people who want to download my product?

Sorry I misread what you wrote in your previous post. My guess, is that blocking the download, doesn't allow Google to see what the download is and since they can't be sure it is safe they prefer to warn users that it might not be.

I am willing to bet Googleweblight appears in a lot of them.

Actually no, this is the first time I bring up Googleweblight and it is strictly the result of my researching the IP addresses you posted.

I am highly experienced and technical.

Did I suggest otherwise?

@glakes Really?

Don't wast your breath or keystrokes in this case trying to explain something to someone who admittedly does not suffer from zombie traffic and has no access to the technical details we have.

We don't agree, get over it. Do you really need start bullying people.

NickMNS suggested that throttling may be a possibility

When did I do that. If remember correctly in the throttling thread I was pretty clear that it was not possible. But again, I don't suffer from Throttling so how would I know, right?

The whole bot theory was retired well over a year ago

Who is talking about bots... The discussion over the last few posts revolves around bizarre zombie like behavior related to traffic coming from specific IP blocks owned by Google. So I guess your right it is zombies, supported by a bunch of hand waving and the "because profits" argument or could be caused by the plausible explanation provided above.

@tangor

Possibly a caution in that g can't see what's there so might suspect the worst and advise their users accordingly? I suppose there's a good reason for blocking g from the product, right?

That's precisely right. If you mess with them, they mess with you. But I had to play with them to find out. As for good reasons for blocking them, they should mind their own business is good reason enough. I wonder whatever happened to the "freedom" you are supposed to have in the US ? Looks like you are all being gobbled up by the big fishes.

Anyway, my problem with NickMNS was that he claimed those google IP's I blocked were operated by indians and brazilians who were able to then shut people down while they use their browsers. I can't imagine hearing anything so dumb in my life.

@glakes

Don't wast your breath or keystrokes in this case trying to explain something to someone who admittedly does not suffer from zombie traffic and has no access to the technical details we have. In the recent past NickMNS has suggested that bots are the cause of zombies

Thanks. I will take your advice. Zombies being bots seems plausible. Is your view that the zombies are actual humans who click through sites ?

As a developer myself, I would implement the zombies in chrome. These zombies can then be directed to access any address on command. That way, zombies can come from genuine chrome users IP's from all over the world. This is in fact what we see happening. So, I suspect google did design the zombies the same way I would. In my world view then the zombies are bots that live inside chrome. Of course I have no proof. But it's a simple and logical solution to the problem of generating somewhat genuine looking traffic that is in fact fake.

@NickMNS

Actually no, this is the first time I bring up Googleweblight and it is strictly the result of my researching the IP addresses you posted.

Googleweblight really has no bearing on any of this. You are not the only one to search those addresses. I had too but found nothing, but I wasn't trying too hard. Please tell where I can find information on those addresses ? A google link will do.

My own experiments tell me those addresses are directly controlled by google. If those addresses are blocked, they will retaliate via the chrome browser. People behind those addresses are also aware of paypal transactions and take actions accordingly. Indians and Brazilians play no part in this.