Google Updates and SERP Changes - April 2013

[sunny_kat]

< continued from [webmasterworld.com...] >

I am seeing a lot of ranking drop. Lot of keywords ranking between 10-50 have disappeared from Google top 100.

Is anyone else also noticing changes?

[edited by: tedster at 11:54 am (utc) on Apr 1, 2013]

[robert76]

Seeing significant ranking changes today with lots more host crowding on page 1 than usual. The host crowding beneficiaries in ecommerce sector are primarily Big A and EMD's. Have seen traffic flux all week but not much in the way of ranking changes until today.

[poweri]

Google is acting strange. When I use site:mysite.com I don't get any results for my new site. Yesterday I saw 4 indexed subdomains, like keyword.mysite.com, and so forth.

But when I search for "mysite.com" (with quotes) I see the 4 indexed subdomains.

I think Google doesn't trust subdomains. It gives no problem when I submit one subdomain. And all pages are indexed within a day. Four subdomains and only the first page get indexed fast and then the strange situation.

[tigger]

Don;t know what happened yesterday evening but something has and I'm seeing a big spike in traffic across a few sites- most of the new traffic coming in seems to be what I would class as low quality but its given me a 100% boost ....will it last ? unlikely

[Robert Charlton]

meta keywords

On a photo site that's all graphics with no text, this fits a pattern I've seen confirmed several times in the past year. When good quality links are pointing to a site, and Google can find no other onpage signals, Google apparently looks at the meta keywords tag (or for any other onpage clue it can find) for confirmation to disable the Google Bomb filter.

See...
Meta keywords - are they worth it?
http://www.webmasterworld.com/google/4471188.htm [webmasterworld.com]

and also...
Meta Keywords still do matter
http://www.webmasterworld.com/google/4490340.htm [webmasterworld.com]

It's the links that rank the site, though... not the keywords or the image file names.

[petehall]

I am absolutely convinced the dial on anchor text just got cranked up on a considerable scale. What I'm worried about is are they trying to encourage over use of keywords right before a new Penguin update?

[indyank]

Google Bomb filter

Robert, what does that mean?

[petehall]

Google apparently looks at the meta keywords tag (or for any other onpage clue it can find) for confirmation to disable the Google Bomb filter.

Which at the moment I'd kind of agree with, however if you remember the original miserable failure example those words weren't anywhere on the page.

@indyank Google bomb is lots of links all using the exact same anchor text to send a page to the top of the results for a particular phrase. A filter would prevent this, as in filter it out.

[luckystrike]

I've experienced the last google panda/penguin first hand having been in the market since 2004 [and enjoying great rankings for my quality sites which now are in the doldrums]

Google present algorithms are just sleazy to say the least. You want the ultimate proof of that? My brother seeing that I was making a successful business tried to copy my business model and setup another site on my subject topic. Now, my brother doesn't know anything about SEO or the subject matter of my site and doesn't care much about copyright so he basically plagiarised other sites heavily to build his superficial site! There is nothing he's written at all there - nothing - and he's already abandoned the project for almost a year now (he didn't build a single link back to the site!) - unlike my flagship site which modestly I believe to be a very good resource on the subject with a monthly magazine and videos and all! [I've worked on it since 2004!]

Today I'm seeing his site ranking 3rd for a competitive term in the industry I'm in [while my site is probably in page 17 suffering the rage of google]. For my flagship site I didn't receive any notifications of manual penalties in google website but I didn't notice a drop in traffic in April 24 2012 - I think even the subsequent Panda affected me. After that my rankings continued to slip slowly but steadily over the subsequent months.

[luckystrike]

small correction: but I didn't notice a drop in traffic in April 24 2012 - I meant but I did notice a drop in traffic in April 24 2012 so I was definitely impacted by penguin.

[JesterMagic]

Ahhh April 24,2012 were I saw 30% drop in traffic. My first major drop in 9 years. I think this is when Google instead of ignoring bad signals applied penalty's for them which affected a lot of older sites. I never used blackhat techniques for link building, and I always tried to deliver quality content. Of course that never stops the competition or who ever from posting the 10's of thousand of links to my site I had at the time using anchor texts of prescription drug terms and sex terms.

[SerpsGuy]

Hey Guys- Just wokeup today April 5th, 2013 and I ran my daily rank check. I am seeing a drop in single word rankings by about 15 places. Drops also occurred on some two and three word terms, less dramatic - 4-5 spots.

This is the change from yesterday. Will report back later today if rankings bounce back or drop further.

[gouri]

I am absolutely convinced the dial on anchor text just got cranked up on a considerable scale. What I'm worried about is are they trying to encourage over use of keywords right before a new Penguin update?

@petehall,

Can you tell me when you say that the dial on anchor text just got cranked up, do you mean that keywords should be used in the anchor text or shouldn't be used a lot?

[SerpsGuy]

My theory is this- Google is alternating between sites allowing them to rank without external links. Then over time requiring more links, meaning those sites that ranked and gained links from it remain. Rinse and repeat.

[petehall]

Can you tell me when you say that the dial on anchor text just got cranked up, do you mean that keywords should be used in the anchor text or shouldn't be used a lot?

It appears to be working well is what I mean... whether you want to head down that road yourself is entirely up to you. I see a site a site sat at the top of page 2 with just over 100 links and over 50% of those are the target phrase.

These words aren't anywhere else within the site; url, text, filenames etc and the site has no pages what so ever. I doubt it will be there long but it's obviously the reason why it's there.

I seem to be seeing an increase across other verticals as well. This is not an isolated result.

[Wilburforce]

Are you sure it's the keywords tag and not anchor text?

My question, also.

[Wilburforce]

I see a site a site sat at the top of page 2 with just over 100 links and over 50% of those are the target phrase.

I'm pretty sure one of my pages has been penalised for exactly that type of backlink profile: the page is about widgets, and a very large number of unsolicited backlins use widgest as anchor-text. The page was at the top of page 1 for widgets (and is still at the top of page 1 in all other engines). Last year it went from the top of page 1 to nowhere on Google. It is a non-commercial (informational) page, and I have done no manipulation or deliberate SEO on it.

[SerpsGuy]

I lost a top spot for the same reason, but that was 2 years ago now. Not relevant info i suppose

[petehall]

I'm pretty sure one of my pages has been penalised for exactly that type of backlink profile: the page is about widgets

I forgot to mention there's absolutely no text on the page... nothing. Not any keyword, let alone the phrase it ranks for.

[Wilburforce]

I forgot to mention there's absolutely no text on the page

That would make a difference!

There are lots of words on my page, but it got clobbered anyway. Perhaps if I had deleted the content...

[backdraft7]

Looks like somebody tripped on the network cable on our server again. NO traffic for long periods today. 20 and 30 minute periods of absolutely NOTHING! Well, at least it can't get any worse that zero? Can it? Come on what gives? I'm still in the serps, but ZERO? Did every on the planet die?

[tumaykilinc]

Google is no longer re-assigning pagerank to domains that have PR N/A for a month now, it usually would take maximum of 10 days. I am talking about domains that actually have pagerank but turned to n/a but still have index and info.

Anyone knows what's up?

[Convergence]

backdraft7

Looks like somebody tripped on the network cable on our server again. NO traffic for long periods today. 20 and 30 minute periods of absolutely NOTHING! Well, at least it can't get any worse that zero? Can it? Come on what gives? I'm still in the serps, but ZERO? Did every on the planet die?

Traffic STOPPED completely 30 minutes or so ago. WOW is all I can say...

[Shai]

Backdraft7, could it be that you are seeing localised organic rankings? What happens if you search for your term from a different location completely using one of the "I search from" type services?

Over the last year or so, we have increasingly found that for some key phrases, you can rank in several cities/towns but not in others.

For example, today we rank number 1 for quite a competitive key phrase that should bring in over 12,000 searches (equates to around 7000 expected visits/month = 233/day) with personalisation turned off, clean browser etc but we are only seeing around ~30/day. If I search for the term from Liverpool/Manchester/Brighton for example, we are on page 4 while searching from Watford, we are number 1.

[getcooking]

@backdraft7

We saw similar patterns yesterday. Not "zero" traffic, but huge drops in traffic at times during the day. Mostly in the afternoon as the morning started off normal (Central time zone). I kept checking the news to see if some disaster had struck that was taking everyone's attention away because like you, our SERPS hadn't changed so it made no sense.

[SnowMan68]

@getcooking

It seems like they have really cranked up the personalized results based on location. In the past when I would check different major cities around the USA we would have pretty consistent results. Now when I check today they are all over the board. I'm talking +/- 5 spots or so which can have a huge impact on search referrals. That may have something to do with your referrals coming in. I would check the major cites to see if you most popular keyword phrases have jumped around too.

[getcooking]

@snowman68

I have a content site so I wouldn't think location would affect us (theoretically).

[backdraft7]

This has nothing to do with personalization. I'm also skeptical it's just Google because I'm not getting traffic from Google, Bing, Yahoo or Google Adwords for long periods of time. I suspect an overall network issue or possibly another DDOS event happening somewhere. It looks like "day parting" because I always get one or two morning sales (3am to 5am) then nothing all day then, like last night BANG it all opens up again for several hours and balances the day out. My web host is also suspect and trying to contact their support used to take minutes or I could call them and get a tech on the phone 24/7, now my tickets go for days and nobody answers the phone. HG in case you're curious. I'm going to check the hosting forums to see if others are seeing the same.

Sat morning...usually banging away, sitting at zero again. The weird part is that I have paid members who are always coming & going. Looks like a network issue.

[TheOptimizationIdiot]

Interesting from The Big G today.

Lately Google's site: page count had been on a roller coaster of +/- 10% every day. Then it went to +/- 15% every few days. Then is went to 3 different numbers throughout the day. (All still indicating well indexed.)

Now, today, they've sent a visitor to a page in a section of the site they show 0 results for when doing a site:example.com/dir search. (It usually works the opposite way, meaning I can find a page using a site: search before I can find it in the results or as the landing page for a query.)

Definitely leads me to believe there almost have to be at least 2 or more distinct datasets out there right now to see those types of fluctuations in numbers for anything. *

* Unless maybe someone's been bored and using their 20% time and breaks to update the site: page estimating system, then revert it a while later just to mess with us, which I guess is a possibility, but that still doesn't account for how they referred a visitor to a page they don't have indexed according to a site: search.

[Convergence]

Our traffic has not recovered. Our busiest category went eight hours without a single visit other than search bots; and only a couple in the time since...

[backdraft7]

I did several key word checks on data sets of 100 phrases each and all show improvement or the same in position. If the phrase is in the top 3 or 4 , they are quite stable while the lower on the page, the more bounce. Even with these figures, the ZERO periods (like a Zombie period without the visits ;^) are quite a mystery. It may be that all of G's April Fools shenanigans upset the data collection process of GA. Not sure I trust these recent numbers, but I will be cross checking again with p wik.