"ZOMBIE TRAFFIC" Separating fact from fiction & emotion

[FishingDad]

This recent discussion about "ZOMBIE TRAFFIC" is just utter nonsense. What are people saying, anything worth while or just a communal <snip> because sales are down on the norm? The talk is firmly in the tin foil hat area.

Are you talking about SERPs, if so why, if your positions are dropping then that's that. If positions not dropping are you seriously saying Google is sending you people they know will not buy from you !? REALLY?!

Are you talking about PAY PER CLICK? if so then your talking possible click fraud then, aren’t you?

Giving any constant period on the internet, people buy or they don't buy and there's many many factors why they will one day and might not the next day.

[edited by: goodroi at 5:55 pm (utc) on Nov 10, 2015]
[edit reason] Let's be careful to keep the discussion on a professional level [/edit]

[superclown2]

is your site in the same place for joe consumer after he has spent 20 minutes searching for the best price or does your site get buried on page 5?

On my main computer I have a habit of deleting cookies and I even change my IP address several times a day to try to get a 'clean' set of results. However I have several computers and I've not seen any major difference in the SERPs on the ones that I don't do this on. It's an interesting line of thought though. Many people never delete cookies so perhaps they are getting a different set of results to me. I certainly get different results according to my location here in the UK - I find London results, for instance, more dominated by big brands than my local results. These differences are not huge but of course one or two places can make a vast difference to conversion rates.

[Jez123]

I am finding that if I don't shut down and restart Chrome (in incognito mode) after EVERY search Google starts to slew the results in my favour when checking my own positions. Try it. Document each one and then shut down after every search and re do all and see if they are different. I presume it's the same for Firefox but not tried it.

[superclown2]

I am finding that if I don't shut down and restart Chrome (in incognito mode) after EVERY search Google starts to slew the results in my favour when checking my own positions. Try it. Document each one and then shut down after every search and re do all and see if they are different. I presume it's the same for Firefox but not tried it.

Presumably you've clicked on your site? Google can of course remember which sites you visited from your IP address as well as your cookies and so weight them higher on subsequent visits; as they have done for a long time now.

Here is a possibility (a hypothesis only, I've no evidence for it); what if a visitors clicks on a high profile 'big brand' site looking for the big widgets that the site is famous for. That same person later types 'little green widgets' into Google which spots the earlier visit and gives extra credit to a different page on the same big brand site even though it doesn't stock, or hardly mentions, the little green variety. It would be a simple matter for Google to vary how much weight to give to this earlier visit which could explain varying SERPs positions, hence varying conversion rates. The webmaster, who has clicked on his own site a lot, or deleted cookies, would be blissfully ignorant of this effect on another computer, in another town.

[Jez123]

Presumably you've clicked on your site? Google can of course remember which sites you visited from your IP address as well as your cookies and so weight them higher on subsequent visits; as they have done for a long time now.

No. I don't click the site. Google still seems to know what I am doing and promote my site further up. On a clean, unsearched, new incog window it's back where I expect it to be.

[Harry]

There is a 15 days delay between the time new articles and pages gain a significant amount of traffic. At first, I thought that it was natural. But the entry and exit traffic does not match the overall views. That means that a series of bots are navigating the site which is a few thousands large and has new pages up every day.

Looking at entry and exit pages only, the old pattern of views and popular pages has remained about the same as before the zombie/ghost traffic.

[edited by: aakk9999 at 2:08 pm (utc) on Sep 19, 2016]
[edit reason] Moved from another location [/edit]

[superclown2]

No. I don't click the site. Google still seems to know what I am doing and promote my site further up. On a clean, unsearched, new incog window it's back where I expect it to be.

Do you delete cookies on shutdown? Does your IP address change?

[Jez123]

No, there should be no cookies collected as in incognito mode. IP address does not change.

[toidi]

When cookies are deleted and a different ip is obtained and in icognito mode, can g still use your wifi to identify you? This is not directed to anyone in particular.

[NickMNS]

The absence of cookies is a signal in its own right. If you search for term without any cookies or history there is a good chance that Google will return results verbatim for the keyword. Whereas, if you have cookies and a history Google will be able to infer an intent that may change the search results.

There is evidence of this occurring, search for a specific term (eg: python) that has multiple meanings, Google will return a variety of results some for the programming language and others about snakes. Once you start clicking on say the snakes, any future (within that session) searches will return results about snakes.

So you may rank well for searches where the user's intent is unknown, but then rank poorly when the user shows a clear intent to purchase.

Presumably, rank checking tools would operate in the domain of unknown intent thus these would reflect the webmaster's experience but not the actual user experience.

[superclown2]

So you may rank well for searches where the user's intent is unknown, but then rank poorly when the user shows a clear intent to purchase.

This makes sense. So, to take it a stage further (whilst stressing that this is all hypothetical) why, to whom, and how, would google redirect the visitor with a buying intention, bearing in mind that we still have a mixture of ads and organics?

[NickMNS]

I theorize that Google must measure conversion rates. It then uses the relative conversion rate of the sites in a particular niche to rank sites. When it perceives the users intent to buy it directs the users to sites with a highest conversion rates.

Conversion can easily be tracked based whether the user returns back to search with a ongoing intent to purchase.

[Simon_H]

@NickMNS Google doesn't need to measure conversion rate, because Google doesn't make money when users convert. It makes money when users click on ads. So theoretically all Google needs to do is identify which users tend to click on ads (very easy to determine) and then ensure those users get more ads put in front of them to ensure their conversion path includes more paid clicks.

[glakes]

Google doesn't need to measure conversion rate, because Google doesn't make money when users convert. It makes money when users click on ads.

Since Google is already collecting this data from goals and for ROAS, CPA ads, Google could use this information to exclude users from a pool of likely buyers if they already made a recent purchase.

So theoretically all Google needs to do is identify which users tend to click on ads (very easy to determine) and then ensure those users get more ads put in front of them to ensure their conversion path includes more paid clicks.

This is the more likely scenario. It's simple for Google to do and produces more clicks and more traffic for webmasters (keeps them somewhat happy). Such loosely applied buyer intent rules may partially explain why some are seeing a horrible ROI with Adwords. With limited knowledge of the inner workings of dynamic bidding, the greatest benefit of it may not be in the advertisers best interest because it gives Google permission to adjust or set keyword bids on the fly. I personally never saw any benefit from dynamic bidding. I was paying for the same zombie traffic no matter what settings I used in Adwords (text and product ads). The only difference being I paid more for the same junk traffic.

[samwest]

After a decent Sunday of almost normal traffic. The week is right back to Zombies.

[samwest]

December 26th apparently Zombie traffic has vanished...for now. Users are now navigating page to page in a more normal fashion rather than sitting on one page forever then leaving. The observation now is erratic streams of ON/OFF traffic patterns, morning only conversions and lots of tire kickers and cart abandons which I consider poor targeting. In most cases you couldn't give FREE stuff away to these visitors.

GART volume varies from hours of "drip traffic" meaning 1 or 2 live users to brief periods of "normal" volume of about 5 to 15 users to sudden onrushes of 50 to 100 visitors on one specific page of the site which quickly diminish to zero again. Hard to believe this would be a natural flow. See throttling topic. [webmasterworld.com...]

[mosxu]

@simon_h

I wonder if your websites are still getting zombies or still getting the mismatched traffic you have described ?

[WhiteHatTryHard]

Yesterday and today I am getting the worst zombie traffic yet:
- Adsense earnings 10% of normal
- Conversions 10% of normal

Huge site with 100s of funnels and articles. Desktop to Mobile ratio unchanged, slight increase in bounce rate (5%). Other than that normal analytics.

I usually get zombies for a day or two, very rarely 3 in a row. Absolutely no idea what causes it. Conversions and Adsense always drop off a cliff together.

However, very often this happens after very small changes to the site design... too small to seem significant for rankings though.

[Simon_H]

@mosxu Sorry, but not really interested in discussing any more. The topic of 'zombie' traffic (as well as much of the WebmasterWorld Google updates thread) has been hijacked by a small number of people who understand little or nothing about the topic, and use it as an excuse to endlessly whine about Google and to 'explain' why their sales have dropped from 3 per week to 2 per week, despite having zero understanding of attribution or conversion path.

Real shame because there certainly are situations when traffic quality from Google varies for a variety of reasons, some obvious, some less so, some innocent, some potentially manipulated, but impossible to debate it on here.

[mosxu]

@simon_h

I really appreciate your insights it is true that some people have less understanding of the conversion path.
Please let us know if John Muller gets back you.

[WhiteHatTryHard]

Of course some websites are not big enough to detect changes that are statistically significant.

But I can assure you that the zombie phenomena is very much a reality, Conversions always drop alongside asense and they usually do so on days with above average visitors.

As to what causes these worthless visitors, I have absolutely no idea.
My best guess is that Google throws different kinds of traffic at you once in awhile to collect data.

Or its some black hat exploit that Google will silently patch one day after years.... it would not be the first time.

[petehall]

Zombies are robots. Zombies are robots. Zombies are robots.

Did I mention the Zombies are robots?

[mosxu]

@petehall

Who's robots?

[moTi]

i can't believe it, this thread with now over 350 posts is about "zombies"? i thought we were professional webmasters. let's get real.

what you people call "zombies" here is nothing but fake traffic bots using headless browsers. it's 2017, spammers prevail, so step up your game.

google has to deal with it (fake ad clicks, fake youtube views)
facebook and twitter have to deal with it (fake accounts, fake messages)

fake, fake, fake..

and guess what? you have to deal with it, too.
no need to mystify this issue at all. use your raw logfiles and analyze your visitors thoroughly. of course bots are getting better every day in mimicking human surfing behavior (that's the purpose) but they still can be detected. get a feeling for which requests are likely from human origin and which must be fake. as you don't want this worthless traffic sucking up bandwidth and ruining your stats, ban it accordingly from your server.

[WhiteHatTryHard]

It it was black hat bots, then you would expect very significant traffic spikes.

Unless Google had caps on traffic to sites, which seems unlikely.

But IF google has some kind of traffic rationing algo, it would be logical to flood your competition with bot traffic. [/tinfoil]

[Shaddows]

I really can't be bothered to engage with this discussion any more either. There are so very few people with useful contributions, along with a small vocal group who fit any faddish discussion into the "Google is evil/broken" narrative. Then you get people who think "I don't see it so it's not real". Who can summon the energy. But before I go...

"Zombies are robots"
"what you people call "zombies" here is nothing but fake traffic bots using headless browsers. "

No. Traffic remains consistent. Unless Google detects the bots on you page and sends correspondingly less organic traffic, at perfectly matching quantities and perfectly matching times, it's just not the case.

There ARE bots, and a bunch of other stuff that is perfectly easy to filter out. Some of us "professional webmasters" can "get real" enough to have clean data to analyse.

I no longer have zombies, they declined for me way before they became a general talking point, and were basically gone before the current zombie threads were started

[mosxu]

I have to agree that zombies may be bots cause normal people in control of they minds and their computers would not say no to a fantastic deal and all of them for hours if not days both organic and adwards.

@WhiteHatTryHard

But bots should be additional traffic to your ads and organic traffic and try to create/buy unlimited traffic to your website and you will see that it will not impact your sales at all. The quota is not on traffic but on the amount of real people in control of their minds and computers..

So, unfortunately need to go back to my theory that the big dogs may be left without enough food because of many like us trying to create great websites and the big dogs will no longer?

[aristotle]

According to the reports here about "zombie" traffic, the logs and stats programs show that it comes from the google search results. However, the vast majority of bots do not claim to come from google. Therefore, the vast majority of bots do not fit the reports about the characteristics of zombies.

As I've said several times, I believe that these so-called zombies are nothing but the ordinary mis-matched traffic that nearly all sites receive. Most webmastres pay no attention to it.

[WhiteHatTryHard]

@aristotle
We can't know for sure what causes it. But chances are it's either Google or BlackHats or both.

[glakes]

As I've said several times, I believe that these so-called zombies are nothing but the ordinary mis-matched traffic that nearly all sites receive.

You have to experience it to fully appreciate the major difference between common mis-matched traffic and zombies. There is nothing normal about paid campaigns suddenly producing a negative ROI beginning in September 2015 while identical campaigns in Bing are still producing the same results as they were years ago. If bandwidth was not so inexpensive, the free traffic Google sends would also produce a negative ROI.

Most webmastres pay no attention to it.

They would if either they were paying for it and/or the saw other search engines outperform conversions from Google on just a small fraction of the traffic. Which is why I say if you don't have a site and its data to evaluate in the zombie pattern then it's nearly impossible for anyone on the outside to grasp. But to give you an example, imagine producing one conversion from 20 organic Bing users and one conversion from 2,500 organic Google users. I have not run the latest data, but I have no doubt its relatively close in comparison.

I don't have any doubt that mis-matched traffic makes up a portion of the dead beat traffic Google sends us most days a week, but there is nothing "normal" about these trends and the stable results we see from other search engines.

[aristotle]

glakes -- If the conversion rate is so terribly low, then the traffic must be mis-matched, or at least very poorly matched. This could happen if the google algorithm goofs and ranks your site for the wrong search terms. For example, you sell "widgets", but google gives you high rankings for "smidgets" instead. A lot of people will come to your site looking for smidgets and won't find them since you don't sell them.

Of course the search results shift around from day to day, so on some days you'll get less mis-matched traffic