Google Updates and SERP Changes - October 2016

[Ukonline1]

---

System: The following 2 messages were cut out of thread at: https://www.webmasterworld.com/google/4817284.htm [webmasterworld.com] by robert_charlton - 3:06 am on Oct 1, 2016 (PDT -8)

---

Starting to get a bit concerned now. Saw very good improvement in serps on 1st September, then fell back to previous levels/slightly worse on 15th September. Seen slight improvement in one keyword and that's it. GSC shows big spike in pages crawled over last couple of days.

I'm starting to think this is how things are going to remain for me now; and after 2 1/2 years of seriously hard work! I really did expect this Penguin update to be good for us!

[westcoast]

"Seeing further slumps today"

Here too. Have seen a steady 1% daily drop daily for the past week and a half.

The main keywords for our site seem to be unchanged. We appear to be losing traffic on long-tail queries. Looks like a lot of less frequently searched for stuff has suddenly gotten filtered out. Very weird, since these tail pages certainly don't have much linking to it and should be completely unaffected by Penguin.

Feels very much like there was some other algorithm change thrown into the mix with site-wide penalties or long-tail changes at the same time as the penguin stuff.

[mboydnv]

Got the same daily traffic numbers here, rankings of #11, yet sales are down 50% for the month.

[RedBar]

Question - Those of you complaining about sales down, are you box shippers or do you have you "unique" products?

[glakes]

Google organic traffic is up substantially (about 60%) and is converting at 1.95%. Adwords shopping campaigns are on and had a 2.94% conversion rate today. Bing/Yahoo organic come in with a conversion rate just under 4.5%. My Amazon sales have remained steady.

[Nutterum]

Google organic is up on my properties as well. Over 20% up from the average m2m and w2w . Conversions have not moved per say, however micro-conversions have increased from 3.4% to 4.8% which is interesting to follow through.

[BushieTop]

@Nutterum and @glakes - had you guys seen penalty. Feels like our slump is because we have a clean history and others are recovering. Not sure how to feel about it to be honest!

[Uber_SEO]

I'm still seeing pages filtered out of the search results. It isn't site wide, but it's page specific. And it's intermittent - one day a page will rank for a given query, the following day it's gone (not in the top 100). Either there's some other algorithmic factors that were launched on the 23rd September, or this new version of Penguin has some very strange side effects.

I'm wondering whether instead of these pages being demoted because of poor quality links pointing at them (we don't have any), could they be potentially being demoted because they're flagged as doing poor quality linking? The noise coming from Google is that filters are being applied at link level right? So maybe if my website has a run of the site link on it to a partner website, then could that cause us to be filtered from the SERPs?

[glakes]

@BushieTOP
Never did anything to get a Penguin penalty nor did any of my competitors from what I can see. We all have very few links. I have been a recipient of Google's zombie traffic though - which some have speculated could be a new modern day penalty. But nothing that should trigger a Panda penalty either - all content is original, detailed and includes unique supporting images/video. Maybe something else was slipped into this Penguin update or Rank Brain has been taken offline for the rollout.

[BushieTop]

@glakes We're bouncing around a lot today! but have seen a slump in the past 5 days. We have absolutely no spam (and i mean literally nothing) going to the page that seems to be in decline. All anchors are either junk anchors, or brand or absolute. no exact match.... starting to think that might be our problem.

[Shepherd]

Main keyword on page 1 for the first time since April 2012. Hope it sticks. Glad we were able to survive/prosper over the last 4+ years despite google's best efforts.

[WhoKnows111]

Yesterday 25% up, today 20% down... not even worth spending time checking it

[westcoast]

"could they be potentially being demoted because they're flagged as doing poor quality linking"

You know, I was having this exact thought yesterday. Our site does relatively heavy external linking in one section, in the way of citations to data / studies / etc on other websites. It adds up to a fair number of external links in bulk, and we are constantly having to keep these links updated.

The part of our site with the most traffic loss over the past week and a half does appear to be those areas with the highest concentration of external outbound links.

From a philosophical standpoint, it does make a whole lot more sense to penalize the SOURCE of poor links than the destination. Sources at least control the links. The old "penalize the destination" was always a bit of a head-scratcher.

It's odd in our case though -- every single external link on our site is NOFOLLOWed, which ought to be as good a signal as any that we're not doing sketchy linking or trying to increase the value of external sites. And our links are merely citations to authority sites wherever possible. Still, "new penguin" does appear to have filtered out or penalized some of the pages on our site with very legitimate, clearly non-spammy, external links (either that, or some other alg changes at the same time did). If this is the case, I can only hope Google tightens this stuff up to filter only true unnatural sources / link farms.

Another possibility perhaps is that links that 404s are now getting penalized more heavily, either intentionally or as a side-effect of the penguin changes. While we do our best to keep up with 404s and repair links, we certainly do have some external 404s in there.

Just throwing some ideas out there.

[EditorialGuy]

t's odd in our case though -- every single external link on our site is NOFOLLOWed, which ought to be as good a signal as any that we're not doing sketchy linking or trying to increase the value of external sites.

I seem to recall Matt Cutts (or maybe it was someone else at Google) questioning the wisdom of blanket "nofollowing" a few years ago. Linking is a basic building block of the Web, and Google owes its birth and success to a link-based formula (PageRank), so one would expect Google to view blanket "nofollowing" as being undesirable: not an offense (I'm not suggesting that), but also not a best practice.

Common sense would suggest that the safest or "best practices" approach would be to link in a natural, organic way--and there's nothing natural or organic about suggesting that you don't trust or won't vouch for any of your outbound links.

[westcoast]

Interesting -- is there any evidence that using NOFOLLOW on a large number of your external links from your site can HURT you?

The reason we have always just stuck them on there is that all of the advice in previous years seemed to be that if you nofollow URLs that you're not 100% sure of, you are protecting yourself from being seen as a link spammer or promoter. Better safe than sorry, and all that.

I can't think of a time that I've heard of someone getting dinged for nofollowing external links?

Note, for instance, that Wikipedia nofollow's all of their external links.

[koan]

westcoast, Wikipedia does that to discourage spammers from manipulating their content to get link juice. Nofollow should be used with user generated content that you don't control and can't trust. If a publisher adds a link on their page as a reference or a source, it has been reviewed and therefore is trusted, so you shouldn't add nofollow. If Google actually gives your site points to linking to quality and relevant sites, you'll miss those points by misusing the nofollow tag.

[samwest]

more speculation or fact?

[westcoast]

"If Google actually gives your site points to linking to quality and relevant sites, you'll miss those points by misusing the nofollow tag."

Curious where you got this notion from? There has certainly been talk over the years that having authoritative links on a page helps that page containing the links (is THAT even a fact?), but I've never heard of NOFOLLOW being a factor in that page boost.

edit to add: I did find this interesting study: [searchenginejournal.com...]
It concluded that outbound authoritive links are a positive factor, but does not answer the question of nofollow vs dofollow outbound links.

[Shaddows]

NoFollow removes the link from the link graph. Vanilla ranking signals cannot pass through NF by design, so you can be neither harmed nor helped by the remote site.

Back in the days where you could better measure individual actions, there was a solid debate if OBLs helped the linking page. My data showed it did, specifically within a site. Others held with the Google orthodoxy that links were votes, and nothing more (i.e. value flowed one way only). This never made sense to me because it implies you cannot get punished by linking to something dodgy.

From an actual business PoV, you need to consider that an OBL is sending your visitor elsewhere, so unless that is an affiliate link (which SHOULD be NF), or unless you have a solid business plan predicated on traffic arbitrage (you get the traffic and send it where it really wants to go, while monetising it when it's there), the ranking boost will almost certainly be more than cancelled out by the outflow of traffic.

[Nutterum]

@bushieTop - Actually I have seen some of the competitors getting penalized. I never had any penalty on any of the websites I have done work on(excluding guinea pig sites that serve the purpose of getting penalized). I think there was a bigger than-Penguin shift in the local SERPs. But that is just me.

[BushieTop]

Seeing a lot of change in the top 5 results today for generic terms. any comments?

[EditorialGuy]

I can't think of a time that I've heard of someone getting dinged for nofollowing external links?

I don't think anyone is suggesting that you might get "dinged" for nofollowing external links. We're just saying that, IF Google takes outbound links into account (e.g., for determining authority), blanket use of "nofollow" won't work in your favor.

Think about it:

Google depends on links for discovery and, to some degree, as "votes."

With that in mind, wouldn't it make sense to take a natural or organic approach to links?

[makera]

Seeing a lot of change in the top 5 results today for generic terms. any comments?

There are still fluctuations in our vertical in top 10 results. I mean jumps from 3 to 8 to 6 etc within 12 hours. Nothing sticks. Wondering if it's going to be a "normal" behavior going forward.

[masterjoe]

I tested this recently on a page with resources on one of my websites... there was a very competitive term ranking at #8, although it doesn't particularly interest me in ranking for it because its not profitable (its informational). However, I decided to see what would happen if I nofollowed all the links on the page, which were linking to medical research journals, quality articles etc, and it then dropped to #18 and down.

The page on my website also has links (followed) from very high profile sites, including wikipedia, huffpost, blogher, and high authority blogs.

[Shai]

Curious where you got this notion from? There has certainly been talk over the years that having authoritative links on a page helps that page containing the links (is THAT even a fact?), but I've never heard of NOFOLLOW being a factor in that page boost.

edit to add: I did find this interesting study: [searchenginejournal.com...]
It concluded that outbound authoritive links are a positive factor, but does not answer the question of nofollow vs dofollow outbound links.

Unfortunately, we were unable to test NOFOLLOW on those links because as soon as the experiment went public, it invalidated any further results due to data contamination. If we kept the study quiet and then added the nofollow, it could be argued that its not the same thing as having the nofollow from the beginning due to theories such as Link Echoes (which we have observed for many years now) [moz.com ]

We had no choice but to separate it into another separate study.

[Robert Charlton]

...there was a solid debate if OBLs helped the linking page. My data showed it did, specifically within a site.

Shaddows, I remember discussion also that anchor text on a page provided a ranking boost. Effects I've seen of insite link boost have suggested that the boost may be related to the quality of pages linked to, and that it's not always limited to literal anchor text vocabulary, but might apply to a collective description of what the links are about. This of course might be influenced by external inbounds to that page... hard to separate that out. I've never done strictly controlled tests where I've tried isolate those factors... if it's even realistically possible to do so.

[Shaddows]

We're an ecom, but we publish loads of case-studies, how-tos etc. One technique that works for us is to publish a set of articles over a month focusing on different aspects of one topic.

We "proved" the interlinking theory by conducting a test, which followed a well established publication pattern, but with minor tweeks.

First we published a single high-quality article on, say, widget-making . It was part of a page-set on hobbyist manufacturing. It ranked well for its term.

On another cycle, the entire page-set was on widget-making. The overview article was much shorter and published first. As we added pages on other aspects, we added links to the overview document, but did not link back. (The daughter pages interlinked in the normal manner). The final page-set was substantially more comprehensive than the single-page article from the hobbyist set, while the overview document was much thinner. Nevertheless, we got the overview page outranking the single-page.

In terms of anchor text, we did also get a strong page ranking for something, say widget storage, ranking for storage materials by linking to thin pages on storage materials. We did not find it a worthwhile exercise in real-world conditions. Better to craft a page to rank properly, and interlink.

Some clarifications on this:
- We are a reseller, and primarily B2B. We do not publish articles on making things, or storing things. The above is strictly for illustration purposes
- The first test was repeatable in real-world conditions, controlling for other factors.
- We have a team of people creating content, who have no other demands on their time. We're good at content.

[Martin Ice Web]

Since the last weekend we see a decline in sales. Traffic was steady so far up until today. Surges of traffic followed by zero traffic times. Sales down to 10%. Something is brewing again as it started just at 12 am.

germany

[highlander888]

UK based eccom site - seen slowly improving rankings across the board this week, edging up a position or two for every phrase every day!. Had very unsettled time earlier in month as rankings have violently oscillated seems to be settling down in a good way. We'll see what next week brings though!

[samwest]

Great days followed by nothing days and complete traffic blackouts.. Same old same old. Does anyone really expect improvements?

[mboydnv]

Tues, wed had great conversions. Did well, like the old days. Thursday meh, fri ok. Today nothing.