Google Updates and SERP Changes - June 2016

[engine]

Continuing from:
Google Updates and SERP Changes - May 2016
https://www.webmasterworld.com/google/4802991.htm [webmasterworld.com]



Monday was a public holiday in the U.S. and also Europe, and I know that many kids in Europe are off school this week, probably on vacation with the parents.

Depending upon your sector, you may be busy, or quiet.

Have you seen the branding tests in Google SERPs?

Put in a search and add the brand name. The brand name appears at the end of the entry, slightly clipping the title.

[edited by: Robert_Charlton at 7:03 pm (utc) on Jun 2, 2016]

[Simon_H]

Geo-matching in the UK still seems to be poor. We're getting significant traffic to a .co.uk domain from Poland and Sweden, all referred from google.pl and google.se respectively. These people are trying to buy, but can't as the site doesn't delivery outside the UK. And, as a user, I'm finding transactional searches regularly seem to show results from the US, Australia, etc page 1. If I type 'buy suv car' from google.co.uk, the autocomplete suggests 'buy suv car', 'buy suv car in india', 'buy suv canada' and 'buy suv calgary', and the results on page 1 and 2 have a lot of US content. Embarrassing.

[RedBar]

And for those of us in the UK we also get to do it again the 6th September, 6/9 :-)

[renatovieira]

@ionguy - My site too. Yesterday was very below average. Looking my Analytics today, things are not going very well too. My site is tourism niche in the USA.

[ionguy]

@redbar
i would suggest to celebrate it everyday to avoid any mistakes. maybe this is some solution to improve positions in serps, traffic, conversions who knows

@renatoviera
i see poor saturday (june 4) and then tuesday (June 7) some abnormal drop in traffic. both days should hit 40% (sat) uq more and 33% (tue) uq more. same things seems to happen today however need wait and see until the end of the day

[samwest]

Every June we see a marginal drop in traffic but now a huge explosion of bounced visits. I'm going to guess kids are out of school and surfing the web, combined with poor targeting, short attention spans and no credit cards.

[aristotle]

samwest -- I would bet that most sites get a lot of mis-matched and poorly-targeted traffic. So what you're seeing is probably no different from what most other webmasters see too. It's just a normal part of owning a site.

[RedBar]

Are those of you seeing less traffic not having responsive sites?

My non-responsive sites have had a dramatic hit since the middle of May mobile update, most are down 50%, whereas responsive sites have either stayed the same or improved, one of them substantially.

[SEOcomfort]

@RedBar I haven't seen less traffic and my site store is not responsive or mobile friendly but my blog is responsive. I have seen pretty stable traffic since mid-May. I believe the main reason I haven't seen a decrease in store traffic is because most of my competitors do not have responsive or mobile friendly sites. Hence, why I am pushing to get our site mobile friendly to beat the competition to what should have been done over a year ago.

[rainborick]

I think that the better question is how many sales are you losing each month because mobile users can't easily browse through your online store. I wouldn't suggest that going mobile-friendly would cause a dramatic increase in sales, but for small sites these days, even single-digit improvements are worth chasing if at all feasible.

[Spiekerooger]

@RedBar:

I've seen non-responsive sites with no ue for mobile visitors rising by 100s of percent in the mobile index ranking since may. Whatever they were doing with the second mobile update, sth. went wrong (with some sites).

But I've seen good mobile ue examples rising as well. Just the number of false positives seems a bit high.

[samwest]

@aristotle - I've been monitoring this particular site for over 15 years. I know what normal bounce and this zombie traffic traffic look like. 10 or more bounced visits in a row? There's something very wrong with that. Site is fully responsive, passes all Google speed tests with high marks. It's just bad traffic...poorly targeted or children.

I just checked my stats for the day (figured I'd work in the yard rather than watch traffic today)...well at 10 am I got a HUGE spike of traffic that lasted for about an hour then slowly trickled off. This has been happening regularly over the past few months. Of course not a single conversion all day even with the spike. Some kind of test is going on...perhaps only on specific sites. Tedster always thought my site was being used as some kind of test site.

Anyone else see these sudden spikes, then decay down to a trickle again?

[aristotle]

well at 10 am I got a HUGE spike of traffic that lasted for about an hour then slowly trickled off.

Could be a tiny DDOS attack

[ecommerceprofit]

* Anyone else see these sudden spikes, then decay down to a trickle again? *

Yes, used to be I could turn ads off for a whole day but now Google is making it tough to predict...on/off happens more rapidly at odd times...the pattern seems more random.

[samwest]

@aristotle, it was not a DDOS attack - DDOS is not human traffic, in this case, it was all traffic originating from Google for individual IP addresses on distinct networks, not from one network. This has happened before, so unless somebody has come up with a way to DDOS a site and make it look like Goofle traffic and have it show up in GA and PIWIK as human, then it was just a surge to a specific page that is related to a higher volume key phrase. But why? I'm still showing rather robust traffic, but no buyers at all.

[aristotle]

samwest -- well referals from google can be faked, and a botnet can include infected machines all over the world.

But you could be right -- maybe it was just a barrage of bad traffic from google. Why don't you check your google search Console (Webmaster Tools) to see if you got a lot of extra traffic from google on that day. If so, you could probably even determine which search terms triggered it.

[Simon_H]

@samwest Yes, try @aristotle's idea in GSC to find the search term(s). Should be pretty easy if it's such a huge spike.

You might then find that there was something on TV/news/newspapers that triggered a load of people to search for that term, your site came up, they clicked it, realised it wasn't the site they were looking for, and bounced.

[samwest]

I know its one sub page that ranks for this higher volume key phrase, but it's "surges" are short lived and infrequent. According to WMT in the last 90 days it occurred on 3/28, 4/18, 5/2, 5/20, 5/28 and yesterday, which has not yet logged. Each time it happens the peak is 10x normal background traffic and always about the same peak volume. It's like I get released from throttled prison every so often...likely accidental or during some kind of refresh. The specific terms and it's close variants appear to be the trigger. So, it's from search queries, not direct traffic. I can only imagine how much better the site would do it this was the rule rather than the exception.

[NickMNS]

@samwest check in the Google analytics,

I have had similar referal spam "attacks" where I get a huge spike 30 to 50 users within a minute. Generally they show up as referral traffic, but I have seen some smaller attacks come through as Google traffic, with search query keyword data (generally the keywords used are completely unrelated).

Typical referral spam signs are:

High number of visits in a very short time from multiple locations/ips all hitting the same page.
Most often the traffic bounces or it will refresh the same page. I have never seen any referral-spam "user" navigate my site. (which makes sense since they are not actually on the site)
The Hostname is not your domain name, often (not set) or something spammy like getfreetraffic.ru

If it is referral spam, then it should have no direct impact on your sites performance. It is simply an annoyance as messes with all your stats.

Hope this helps.

[aristotle]

As I've mentioned before, owners of ecommerce sites should forget about zombie traffic, mis-matched traffic, referal spam, fake traffic, rogue bots, etc.

The real issue for ecommerce sites is what google is doing with people who know what they want and are ready to buy. Call these "likely buyers". I believe that google can identify most of these "likely buyers" by the search term that they use, their search history, their location, and perhaps other factors. I also believe that google gives special attention to these likely buyers and probably shows them a special set of search results.

So it's these likely buyers that you should be thinking about, not zombies etc

[ionguy]

@aristotle
this is the best hypothesis i ever heard - the funny thing is that you may have right

[mrengine]

So it's these likely buyers that you should be thinking about, not zombies etc

On days when there is good buyer traffic, zombies do not exist. On days with weak buyer traffic exists zombies herd my site and level out daily traffic to strangely match the days when traffic is productive. Zombies are almost like an artificial filler for my daily unique visitor logs. Whether one focuses on figuring out where likely buyers are going or why zombies are coming will answer the same question.

[samwest]

Agreed Aristotle. I've always suspected that scenario. Google just recently announced that you can search for yourself now and get everything they know about you, and I'm sure for some, that's a lot!. Keep in mind, by using their apps, we give them permission to listen in on our cell phones while having a casual conversation in the privacy of your own home. I believe this snooping and selectivity began evolving somewhere about MayDay 2010, because it's been all downhill or sideways since.

One more related thought, I ordered a new motorcycle helmet the other day, it was a specific brand and style. The day after, my wife began getting targeted ads on FB for ladies version of the same model. Her first thought was "are they listening to our conversations?", maybe not, but with banking data and our traffic habits they have more than enough data to target us all.

[edited by: samwest at 8:48 pm (utc) on Jun 9, 2016]

[NickMNS]

@samwest it could be even simpler than that, did your wife post anything about your new helmet on the her fb timeline?

[samwest]

Nick - absolutely not...in fact, she could care less since she's a nurse and won't even get on the bike. We don't post to FB very often and if we do, it's certainly not to gloat about junk we bought.

[NickMNS]

Scary!

At Christmas time I tell my wife, when using our shared ipad to surf for gifts using duckduckgo otherwise I know from the ads what she's planning to get me as a gift.

[Shepherd]

I believe that google can identify most of these "likely buyers"

this is the best hypothesis i ever heard

This is no hypothesis, google flat out claims that they can do this, in fact you can set up adwords to increase your bid for searchers that google has identified as having a high probability of making a purchase.

[Robert Charlton]

...a HUGE spike of traffic that lasted for about an hour then slowly trickled off

Try this WebmasterWorld site search...
[traffic spikes site:webmasterworld.com]

One Hour Traffic Spikes
Jul 1, 2013
https://www.webmasterworld.com/google/4589510.htm [webmasterworld.com]

I've noticed over the past few weeks that several of my (I believe) Panda/Penguin affected sites seem to receive a massive rush of traffic during a given hour in the early morning....

Also, for more discussion on the above, see...

Seeing some big traffic dips with no ranking changes
Nov 27, 2013
https://www.webmasterworld.com/google/4626454.htm [webmasterworld.com]

[Robert Charlton]

when using our shared ipad to surf for gifts using duckduckgo otherwise I know from the ads what she's planning to get me as a gift.

I'm not sure that the search engine used matters. Ecommerce sites, eg, use "retargeting" to make sure that visitors who've looked at their sites see additional or related ads. ISPs and ad networks also sell this data. I'm still seeing ads for a small appliance I bought a month ago.

[NickMNS]

True, but it is not as obvious as when using Google directly.

[masterjoe]

I've been having some fairly average traffic, zombies have been at bay for a week or so now. However, since this rubbish started happening I immediately put more effort into a/b testing an opt in form and offer, so on zombie days I can send out an email and somewhat make up for lost sales. This has helped considerably, although there are still noticeable dips in conversions every now and then.