Google Updates and SERP Changes - September 2016

[NickMNS]

---

System: The following 4 messages were cut out of thread at: https://www.webmasterworld.com/google/4814310.htm [webmasterworld.com] by robert_charlton - 3:01 pm on Sep 1, 2016 (PDT -8)

---

I agree with you that features such keyword, location, browser type and so on (not set) may well be someone blocking the info for privacy reasons.But @ionguy my understanding is that whenever the hostname is set to anything other than your domain name then it must be a bot, this includes (not set). In so far as hostname specifically is concerned (not set) is always a bot and not a privacy issue. Filtering out the hostname (not set)s and other foreign domain from your data is essential to getting accurate stats.

This is applies to piwik and other analytics tools as much as it does to GA.

[liamkk]

@hopepro

Crappy sites get a very brief improvement before being hit with a penalty. This is because Google does a deep crawl on 'sites of interest' and a fresh crawl always improves visibility and rankings for pages.

I wish you all the best.

[Nutterum]

OK a big update. Over 100 local business websites got eviscerated from the SERPS. Allegedly most were worked by 2 or 3 agencies but still... I have not seen a hammer drop this big since Panda 2. Local Australia.

[BushieTop]

@hopepro, try not to worry - the same has happened to me and i believe others too.

[Shai]

Liamkk, Thanks for the clarification.

My point is that you made a statement of fact which is based on no real data or experimentation. You said:

"Website speed has no impact on ranking."

I'm just saying that unless you have ran some good quality experiments on it, controlled as much as the environment will allow, you just cant say that in that level of certainty. I agree that it is incredibly unlikely that Google will penalise for browser speed, but repeated slow response times during crawl, peppered with a few 500 server unreachable errors and maybe even the neighbourhood the site is on (a shared server with 1000 other sites) can paint an overall picture of the trust the site should gain. This in not about fair or 100% watertight rules. This is AI trying to find patterns.

If your site, which is clearly lean and fast avoided any effects in this update then it just strengthen my feeling that UX had something to do with this update. I don't think there is any harm in trying to gain some data from the people here that had the worst drops on this to see if we can find a clear pattern. It may not be the only component but it may well be one of them.

[BushieTop]

Speed does have an impact on ranking, especially Mobile IMO.

Not the thread for it, but there's a definitive correlation between when we drop ranking and low converting traffic. Today is horrible.

[Robert Charlton]

There's a very split opinion on the CTR vs ranking thing. The studies are quite flawed.

Extremely flawed, and CTR and bounce rate, in particular, have been discussed ad infinitum. Last good thread I can remember here was this one, worth a read...

Click-through Rate, Pageviews, and Time on Site
Nov 2015
https://www.webmasterworld.com/google/4776413.htm [webmasterworld.com]

Google has maintained, and I agree, that CTR is a noisy signal, and that the metric can easily be spammed. Rand Fishkin has been the driver behind most of the "experiments" on CTR, using a conference session or a video blog presentation to induce viral clicking, seeing if that will boost a listing in the SERPs. The best that can be said is that some slight ranking movements up for short bursts have been observed, and that the effects don't last very long.

I think the best theory I've seen about this movement is from Eric Enge, probably describing Google Trends kicking in when anticipating a breaking news story. In a video with Google's Andrey Lipattsev in conversation with Rand Fishkin, Ammon Johns, and Eric... I remember that Rand was fairly pushy in injecting his CTR theories into the discussion....

In a follow-up to the video in this blog article, Eric reiterates his interpretation....

Why CTR Is(n’t) a Ranking Factor
27 April 2016 by Eric Enge
[stonetemple.com...]

...I'll stand by my guess that I made in the [video] conversation that some part of the Google algos that are designed to pick up on hot news events is triggering the behavior seen in Rand's experiments. This would explain the rise in the results and the drop afterwards when the click activity tapered off. But, we can’t know 100% for sure.

I don't think, btw, that clicks in Google Ads themselves would have this effect... but I can certainly see that if a site caught on because of Google ad traffic, links might follow and that could raise rankings, which might explain other observations in that thread.

---

More to come, if I can manage the time, on cannibalization, and also other aspect of SEO as I see it, beyond rankings, which might improve conversions.

These are topics this forum simply hasn't been discussing... topics like landing page optimization, conversion optimization, ehancing UX, etc, which many small sites simply haven't done, but you can bet that Amazon, eg, certainly has. (And yes, they do cost money and/or take time).

The positive effects of Amazon's conversion optimization efforts could easily be mistaken for Google sending Amazon better converting traffic. IMO, its more likely that Amazon is doing all of the heavy lifting. Google is just following what its users want, in part because a good user experience is also what Google wants. Rather than knocking Google for it, I'm impressed that Google appears to be fair with Amazon, sticking to the algorithms and not trying to suppress a competitor. Yes, Amazon and Google are competitors on many levels.

Worth noting I have many of the posters in this Updates and SERPs thread post only in this thread and in AdSense. I've never seen some of them post about any other topics except rankings and earnings.

[Simon_H]

@Robert_Charlton I can't make an educated call on CTR vs rankings, other than to say that the studies seem flawed. But there does seem to be at least some agreement even in the links you sent that there may be a temporary effect when big changes in CTR happen until Google readjusts. The change to the images widget (and also to the local pack and maps) that happened over the last day or so would certainly impact CTR of organic results in the same SERPs. Whether or not those CTR changes would be major enough to temporarily upset rankings (assuming there is a temporary relationship) is unclear. But either way, I do think it odd that Google should make a major change to the images widget and also the local pack and map at the same time things seemed to be going crazy.

[Shai]

@Robert. I agree that the rankings changes are temporary but I do not agree they are slight. We had an article go viral (and I do mean viral, it even reached India, Australia and Vietnam) and the rankings went from 9 to 1 for an incredibly competitive key phrase. This is common when you do viral marketing properly and I can show you some brilliant examples of this. This type of increase would normally require a good handful of very juicy, hard to come by, links. However, as you say, and as Rand Fishkin observed, it was short lasting, dropped slowly and gradually almost exactly as the traffic was decreasing as the story ran its course and fizzled out.

Thanks for the thread links. Will definitely give them a read today.

[hopepro]

@liamkk

Thanks for this, this is really new for me

I also notice that Google Crawls my site with wider interval. Normally, new post would index within 1-2 minutes (using wordpress) but now it will only index all posted content during my night time (I tried submit the post in google webmaster tool it doesn't work either the post appears in search for a few seconds and it's gone, it appears again with auto indexing next morning). There's another section of my site that's using framework as main CMS and I also notice that it takes Google about a day to index the new content too.

@BushieTop

I would say I'm really panic with this tsunami waves which keeps on and on. thanks for your nice advice

[syedyasir]

Today, I am seeing some improvement/recovery in my site's ranking, but overall, the ranking is still down. I hope ranking is going to improve further...

[Nutterum]

@Shai - I managed to create a viral blog post around Brexit and got over 4000% increase in unrelated organic traffic for 2 days. Pretty much I was dominating the entire Topic Pillar. Within a week the traffic went back down to it's normal levels. Unless you are a huge authority that has a small army of content creators, you will have no chance of riding the "virality" wave on the SERPs.

[Robert Charlton]

I agree that the rankings changes are temporary but I do not agree they are slight. We had an article go viral (and I do mean viral, it even reached India, Australia and Vietnam) and the rankings went from 9 to 1 for an incredibly competitive key phrase. This is common when you do viral marketing properly and I can show you some brilliant examples of this. This type of increase would normally require a good handful of very juicy, hard to come by, links.

Shai, I'm not disputing the effects that a genuinely viral article can achieve, but, as I assume and you describe, in a viral campaign, much more than simple CTR is involved... Those good solid links coming from a pattern of clicks is essential. I'm thinking also that the whole campaign has a pattern resembling contagion.

When the buzz and viral effect are real, I'm virtually certain that Google can watch the effect of a real viral campaign spreading, and that the nature of the propagation, and the numbers, and the searches about it as it happens, are of a different kind from the patterns coming from a test.

In a real viral campaign, I also often notice commentary about the viral phenomenon itself... and that the publications writing about the piece are, within a niche, legit and respected.

Examples I'll pull out from memory, one from the past... two more recent... that you might be familiar with...
From some years back, "United Breaks Guitars" (in which Rand, I'm guessing, was involved, at least in the social campaign). More recently, the Michael Phelps Under Armour "Rule Yourself" spot that peaked during the Olympics was brilliantly executed... and, still building... the XKCD Climate Change cartoon. These are all short pieces, several involving video, so not exactly the same as an article, but I assume the nature of the spread was similar.

I think this differs a lot from the placement of widgets and local paks, which just feel different to me than something viral. It's not the same realm.

[Shai]

Thanks for the "United Breaks Guitars" reference. I never seen that before. Really made my day. Great stuff.

Nutterum - Yep, that's exactly my experience.

[mosxu]

if Google algorithms are designed to spot hot news, CTR on the up so they can spot CTR going down. Imagine a competitor ranking for your brand terms where you have most of the CTR and CTR manipulators click these terms more than you usually get from normal users the result will be a drop in general rankings and junk traffic but the competitor will move up in the rankings as Google concludes that competitor brand is stronger than yours since is beating you at your own brand terms.

It happens a lot in casino/betting where competition is a lot higher.

[30K_a_month]

@shah" "If a site is very spammy, it does not matter how much 'authority; you think it has, it is always at risk of going down as soon as Google catch up with it"

No true, im my experience and because of negative SEO. no matter how many bad links your fire at some sites they will not move downwards. Your information is false.

@shah "As you have no idea what information we hold, I would say your last sentence is an assumption. Obviously our observations are just that, observations and theories but I'm trying to base them on real facts. As opposed to what you seem to believe, we hold quite a lot of data on quite a lot of sites."

Wow, you said I made an assumtion then assumed you had more information than me. Classic stuff.

Anyway apparently you have concluded that this is not about links/penguin. Lets keep that in mind as this develops.

[NickMNS]

I'll share some anecdotal evidence about CTR impact.

First some background, my website is large, I provide statistics about widget, millions of widgets. The more popular widgets have more relevant statistics, where as the less common widget has only a few statistics. My site tends not rank so well for the more popular widgets because others have covered the topic before and have a better link profile. I tend to do well with widget that are not that popular but for which I have a fair amount of info, at for which there is some search volume.

I do not rank for any particular keyword but for I range of long tailed keywords. As a result users land on a different page each time. On any given day I generally the most landings on a single page tends be in the order of 10 or 12. That pages is rarely the same day to day. I am excluding from this my home page and other category pages.

During the Olympics as result of medal being one by a US athlete I had an instant surge in landings on the page of one particular widget. The widget in question had very little information about it, a nothing directly related to the searches intent. In the span of 1 hour, I had received well over 100 hits to this particular page. The bounce rate, was over 95%. Pogo-a-gogo.

I generally have a bounce rate in the order of 80%, this is due to there being a lot of relevant information on one page and the information on any other is more detailed and does not interest the majority of users. So going, from 80 to 95, in addition to my assessment of how well the intent search terms are addressed leads me to conclude that these user truly bounced.

So this describes a perfect storm, a semi-viral in-flow of users that all bounce. If what is described in terms of CTR was true I should have seen a drop in traffic. But that wasn't the case I had no statistically significant change in traffic patterns and the general slow growth trend continued.

Does CTR have an impact on rankings, maybe. But one thing is certain it is not a as simple as it is made out to be.

@Simon_H IMO when it comes to removal/demotion of image search, the CTR impact is not required to cause the flux. Simply moving sites up, and moving some from page 2 to page 1 maybe sufficient for the webmasters to feel the difference. It was certainly enough to cause Mozcast metrics to go crazy.

[Simon_H]

@NickMNS Removal/demotion of the images widget would certainly cause a change in *traffic* as different organic results would be higher up the page, but it surely wouldn't change *rankings*. So, for example, #1 would still be #1 even though the images widget was now below it rather than above it. For the actual rankings to change, something else would need to be at play. No?

[Shai]

@30K_a_month

"no matter how many bad links your fire at some sites they will not move downwards. Your information is false"

Now you are talking about a quality site that has had spammy links directed at it (such as in the case you give, Negative SEO). That's not what we originally discussed. You said that a spammy site, if its there for long enough, is somehow immune from a Google penguin update or some sort of link penalty.

"If spammy competitor is still there and has always been there then hes already got enough authority"

I agree that some sites can reach quality threshold that would make Negative SEO pretty much impossible. For example, it does not matter how many crappy links you point at Microsoft.com, it will not stop it from ranking. But that's not what it sounded like you said. Did I understand you incorrectly?

[NickMNS]

What does it mean to be #1 given that today you have images, knowledge graph, local search, ads on top, ads on the bottom, 1 ad, 2 ads , 3 ads, 4 ads. More so given that all these factor are continuously changing.

So my question to you is how do you determine your rank? And if you are ranked #10 and appear on page 1 of the serps then is it the same as being #10 but showing up on page 2?

[aristotle]

Here's a strange example of the effects of a news event. In 2007 there was a mass shooting at Virginia Tech university, with more than 30 people killed.

One of my sites has an outlink to one of the departments at Virginia Tech university. My site, and the page I linked to, have absolutely nothing to do with mass shootings or any kind of violence. There's no relevance at all -- none.

Yet the page that has that link received more than 1000 referals from google, yahoo, etc on the day of the mass shooting and subsequent days. Think about that.

[Shai]

Fascinating. Any idea what the search terms were from some of those referrals?

[aristotle]

Shai -- The main search terms were "Virginia Tech", "Va Tech" and "virginia Tech university". My site doesn't have any information except for the anchor text in the link: "Virginia Tech Department of ____________", but that was apparently enough

[Simon_H]

@NickMNS Absolutely! I'm constantly saying that looking at any part of the SERPs in isolation is largely pointless. Being #1 in organic means anything from being at the top of the page, to being under images, local, shopping, 4 search ads, etc.

Yet, despite this, rank trackers and lots of SEOs still think of everything in terms of rank, i.e. your position in the organic results. They get confused when traffic changes and rank doesn't. Many of the changes that people assume are organic algo updates could just as easily be something like Google tweaking the thresholds on when the Shopping widget is shown.

[Robert Charlton]

if Google algorithms are designed to spot hot news...

Regarding the QDF (Query Deserves Freshness) boost, here's another thread that might provide interesting reading...

June 17 Google algorithm changes: "News-Wave Update"
June 22, 2015
https://www.webmasterworld.com/google/4753914.htm [webmasterworld.com]

Additionally, as some members have noted in various algo discussions (the Google algorithm being the core topic of this Google SEO News forum), the way Google looks at certain queries has in fact changed.

Broad, general queries are the most likely to be hit, but some local queries, those that might be affected by hyperlocal mobile, have also been in effect "redefined". I tried to bring that up in this thread, unfortunately too late to get any serious discussion. Note my last comments in the thread, about "needs met"....

Google Quality Rater Guidelines Update March 28
April 2016 - July 2016
https://www.webmasterworld.com/google/4799052.htm [webmasterworld.com]

[Shai]

Missed this yesterday. John mueller says "fluctuations in search are normal"

[searchengineland.com...]

[liamkk]

It looks like similar to Panda - we had an influx of Google robot activity crawling old 404 links from first Panda days. They were long dropped from the webmaster tool but crawl bot has checked them again.

[Simon_H]

Something which no-one seems to have mentioned... at the same time (~13th/14th Sep) that there was a huge drop in the images widget being shown or it being demoted, the SERP count jumped right up according to Mozcast.

So it seems that the images widget was *replaced* with more organic results on page 1, not just removed. That's also going to impact traffic, especially if you were previously at the top of page 2 or bottom of page 1.

[highlander888]

I noticed something unusual.

One of my sites design was changed 31st August, I checked the cache regularly and 2nd Sept Google had replaced the old cache with the new design.

I checked back on Thursday and the cache date showed 9th Sept, but the design show was the old one.

It updated again yesterday, back to the new design and with a cache date of 15th Sept.

Some of the main internal pages have now switched back to cache dates of July.

The site has 400 pages and according to the Search Console Google crawls around 200 per day, which has been the historic average for many years.

Certainly seems something is switching backwards and forwards.

[sandboxsam]

Hi liamkk, I also saw 1000's of 404 links show up in Google Webmaster Tools (GWT) on 9/12 for old pages that are long gone.
Also saw several crawl spikes by Google starting on 9/11. Still continuing as of my last GWT report on 9/16.

[masterjoe]

If you're all over it, I understand. But I have been having huge issues with zombies for the last few weeks. After deleting my campaigns, I decided to just test a handful of keywords that I KNOW convert... except that they didn't and happened to have drained a $50 budget without a single conversion. These terms used to convert well organically just earlier this year, but things seem vastly different than it used to be. Conversions are lower even though I have tweaked certain campaigns significantly over months at a time.