Google Updates and SERP Changes - August 2014

[webcentric]

---

Continued from July 2014 thread at: http://www.webmasterworld.com/google/4684131.htm [webmasterworld.com] by robert_charlton - 10:43 am on Aug 8, 2014 (PST -8)

---

i think to mention that traffic is down and some other WM agree to it will help webmasters see that it is Google changing the system again and not the a Problem with the site

My point was simply that changes do not impact all websites at the same time, generally, but instead appear to roll out in phases across topics, site-types and/or localities. As others have pointed out, if there are 500+ algorithm updates/tweaks/whatever every year, that's over one per day. Question is, is the current tweak rolling out in your neighborhood on the same day (or even the same month) it rolls out in mine.

All I'm saying is that if the changes were more carefully monitored here by location/niche/etc., it might be easier to see what is actually changing at any given time and the scope to which that change applies.

It's pretty apparent that not everyone experiences the same algorithm changes at the same time or in the same way. Maybe there's a regional aspect to some of this e.g. a change rolls out in America today so expect it in the UK a week from now or whatever. I think there needs to be an effort to differentiate what's happening in this niche or in this neck of the woods from other places and/or topics.

It takes time to rebuild indexes the size of Google's and it's expensive which should explain index shuffles that only impact small segments of the index. Test it here today, if it seems to be right, roll it out elsewhere tomorrow. What's happening in the US today may be completely irrelevant in the UK at the moment. What anyone sees at any given moment could be a pigeon in the coal mine (a warning of things to come) or a red herring (just an experiment that never gets fully implemented).

One can choose to track it in a scientific fashion or just site back and watch. Not sure one path is better than the other but if you want to make sense of it, then applying a little differentiation to the observation process seems like a sensible approach. Bird watchers can count birds 1,2,3... or they can identify them by type and then count them (birds of prey, songbirds, migratory birds, etc.) Simply counting them doesn't seem that useful to me but I'm no bird expert.
.

[edited by: Robert_Charlton at 6:56 pm (utc) on Aug 8, 2014]

[RedBar]

I totally agree with you and it is plainly obvious that "whatever" may be happening it is good to have canaries as a possible warning system insofar as Google is concerned.

[Wilburforce]

canaries as a possible warning system

e.g. Mozcast (spike on Wednesday): [mozcast.com ]
.

[edited by: Robert_Charlton at 6:50 pm (utc) on Aug 8, 2014]
[edit reason] splicing update threads [/edit]

[nomis5]

Mobile usage now accounts for 25% of all web usage across the globe - up 14% from a year ago

Yep, until they define what their use of the word "mobile" actually means, that phrase means sweet fanny adams.

And it's interesting that after all these years we don't know what G means by "mobile". Strikes me that the glibness of the word could be useful for G, to some it means "this", to others it means "that". Who can ever prove them wrong or right.

[EditorialGuy]

And it's interesting that after all these years we don't know what G means by "mobile".

Trouble is, any definition of "mobile" is likely to change over time, and no definition is universally accepted.

Is a tablet "mobile"? Where do "phablets" fit into the picture? Have any studies emerged that show different mobile-vs-fixed-base usage patterns for, say, small tablets like the Nexus 7 or the iPad Mini and full-size tablets? (I remember one study a few years ago that said most tablet usage was at home, but it's easy to imagine that a 7" tablet like the Nexus 7 might be carried around in a purse, a handbag, or even cargo pants.)

Still, it's true that Google's signals can be confusing. In Google Analytics, Audience/Mobile/Overview has three categories (Desktop, Mobile, and Tablet), but Audience/Mobile/Devices is a single catch-all category for tablets and phones. And that's just Google Analytics, which may categorize things differently than the Search team does.

[nomis5]

Still, it's true that Google's signals can be confusing. In Google Analytics, Audience/Mobile/Overview has three categories (Desktop, Mobile, and Tablet), but Audience/Mobile/Devices is a single catch-all category for tablets and phones. And that's just Google Analytics, which may categorize things differently than the Search team does.

Yes, I was just checking up on that in Analytics and found exactly what you say. The thing is though, it's a crucial distinction for us and clearly also for G.

What confuses the situation even more is that it is probably hugely, geographically skewed as well. "Mobile" usage is probably far more prevalent in less developed world countries (no definition of that either) than it is in Western Europe, Amarica, Canada, Australia etc. The statemment that 25% of internet access is via mobiles globally may in truth be a red herring for many website owners.

I trust my stats for my sites and I see a split of 50% desktop, 35% tablet and 15% mobile (G's analytics definition). So from my perspective the news is not that mobiles are 25% it's that tablets are 35%.

On that basis (my sites look fine on tablets just they are) all the responsive design stuff I've gone through has not brought the benefits I expected.

I also think that tablets (maybe even phablets) are the way the internet will ever more increasingly be accessed in the future. And if that happens I'll be right pissed off that G put me through all the trauma of responsive design. I certainly bought into it but I'm not so sure now.

[EditorialGuy]

I trust my stats for my sites and I see a split of 50% desktop, 35% tablet and 15% mobile (G's analytics definition). So from my perspective the news is not that mobiles are 25% it's that tablets are 35%.

For our main site, "mobile" now outranks "tablets" as a source of traffic. That came as a surprise to me when I discovered it this morning after reviewing our GA statistics for July--especially since only about 200 of our 5,000+ pages are optimized for mobile users.

[RedBar]

What confuses the situation even more is that it is probably hugely, geographically skewed as well. "Mobile" usage is probably far more prevalent in less developed world countries

Now there's a can of worms!

I travel far too much on business however I know for a fact that mobile is huge in both China and India, in fact in China they're no longer laying physical lines since it's cheaper and faster to construct mobile networks and much easier to repair after severe weather conditions or disastrous earthquakes etc.

The largest mobile operator is China Telecom with some 775 million subscribers and in 4th place India's Airtel with 275 million subscribers and then the question has to be what are "less well-developed world countries"?

I travel to these countries regularly and I can tell you that a lot of stuff is way, way ahead of the UK and USA.

[EditorialGuy]

I know for a fact that mobile is huge in both China and India, in fact in China they're no longer laying physical lines since it's cheaper and faster to construct mobile networks and much easier to repair after severe weather conditions or disastrous earthquakes etc.

That certainly blurs the definition of "mobile." Does "mobile" refer to devices or networks? (In some countries--even in Europe--quite a few people use cellular USB devices on their laptops and desktops. Are they counted as "mobile users" or "desktop users"? The answer depends on who's counting and what's being counted.)

[Dymero]

Took a look at a bunch of SERPs for the first time in a while and am seeing a disconcerting number of nine-result SERPs.

Ecom, US

Edit: According to Mozcast, the average result count went down a bit on Wednesday.

[nomis5]

quite a few people use cellular USB devices on their laptops and desktops

I'm from the UK and that's me exactly! I have no alternative living out in the middle of nowhere. The whole village does the same as do the three nearest ones to us.

If you can get good reception it works passably well. And how we are counted I have no idea - mobile, laptop, stupid idiots for living in the countryside? Who knows.

[graeme_p]

"Mobile" usage is probably far more prevalent in less developed world countries (no definition of that either)

Yes and No (assuming where I am is a fair sample). More people use mobile, but they tend to be light users which means their impact on usage numbers in terms of pages viewed or time spent is lower.

[trabis]

Algoroo is on the red right now (August 7th).

Algoroo reports a high level of turbulence in Google’s search results after more than 2 months of silence

Update: A review by our chief engineer reveals no obvious anomalies with our dataset. We’re probably looking at a genuine algorithm update at Google.

Perhaps a Penguin or Panda update is brewing?

I did not notice fluctuation on my side and it is a shining day according to mozcast.

PS: Serp Metrics is on fire too.

[chalkywhite]

Had a crazy spidering session two days ago. Spidered 1900 pages. normally around 300 per day. Sitting waiting to see what it was.

[samwest]

I concur with the above observations. Seeing a huge drop in traffic volume and conversions, but spidering is going ballistic!

[Itanium]

maybe the weather

[samwest]

it's not the weather...we've taken a beating by Google in the past, but zero people on the site for hours at a time? Look like we're done. I figure G has to boost it's quarterly billions, so all traffic is going to ad network partners.

[ibeau]

My Impressions have been abnormally high over the last few months (while clicks have stayed the same). However over the last day or two since this latest flux, my impressions have settled back down to "normal" and traffic has increased.

[Robert Charlton]

samwest, sorry abt your problems. Is this the same site you're asking about in this thread, where you had the sitemap/https error?... [webmasterworld.com...]

[Martin Ice Web]

I join samwest, with big drops, hours with -95% traffic with folloowed spurts of traffic. Overalltraffic is at about 60%. Bounce is near 100%.
And yes we changed a little bit the navigation but this is about 8 weeks ago.
I know of two others seeing the same. It feels like a new panda.

[Kiran84]

Same here, no warning in wmt and no changes made to site.
organic traffic nuked.

[i.imgur.com...]

[Martin Ice Web]

Kiran, ecom or informational?

[Kiran84]

regional forum with few million posts.

[Martin Ice Web]

kiran, do you have a login form?

[Martin Ice Web]

kiran, do you have a login form?

in our niche we see the biggest changes since panda was introduced.
Again brands ( amzon for sure ) got a big boost and occupy #1-#6
Some of this brands don´t even offer the searched item. Sometimes domain crowding is back up to 5 positions in a row.

Followed by mostly unrelated nonsens shop/info sites. Site i would never ever buy from! Sites with mega menues and only 3 word descriptions or copied manufactures description.

It seems that smaller ecoms got hit again. I don´t know if their EAT quality guidleines are only a joke but this is afull to see.

[samwest]

wow! I though my drop was bad, impression here dropped from 4,000/day to 1,000/day in 3 days. With 200 measly pages of content on my site, I wonder how anyone could possibly generate a "few million posts"! Kudos to you!

BTW, other sites I administer are showing identical drops in search queries across the board.

[EditorialGuy]

Our main site's Google organic traffic has been creeping up over the last few weeks, but yesterday (Sunday, August 10) brought a spike of about 10%.

We had our first deep crawl in quite a while on August 5, and we usually see changes about five days after a deep crawl, so it wouldn't be a surprise if some kind of update or refresh affected us yesterday.

[RedBar]

My gut feeling is that something is happening going by some deep crawls however an entire swathe of my European trade is now on annual leave therefore it's difficult to judge what.

Interestingly my main.com does not appear to have been affected as yet.

[Broadway]

I've had trouble with Panda since 2.0. I had a significant partial recovery with 4.0.

The last few hours Analytics has shown a significant drop in traffic (on the order of over 50% as compared to the previous week) for my main site. I fear its the update mentioned here that's rolling around to me.

[Wilburforce]

We had our first deep crawl in quite a while on August 5

I had the biggest crawl in a very long time on 9 August.

In the last few weeks there have been several days - another yesterday - when I have had almost no referral phrase information in my server logs (with normal levels of traffic).

It looks to me as if this is being blocked by Google (probably for a few hours at a time on specific servers).

Is anyone else seeing something similar?