(39)
As I reported last night out of the SMX Advanced keynote with Matt Cutts - the third version of the Google Spam Algorithm, also known as the PayDay Loan algorithm, is launching today.
Didn't we just have version 2.0 a couple weeks ago? Yes we did.
Matt Cutts explained that 2.0 was targeting specifically spammy sites, while 3.0 targets more spammy queries. So the algorithm looks more at the query versus looking at the site. You are smart SEOs, so you can figure out what that means.
Matt also explained that version 2.0 also added some negative SEO factors, to reduce the amount of negative SEO that can happen. If you believe that.
I do not believe this launched just yet, it may launch today sometime, or maybe tomorrow - but likely today, Matt Cutts said. Based on the lack of complaints in the black hat forums, I do not think it launched at the time I am writing this.
Reminder, this will look specifically spammy queries, such as terms like [payday loans], [casinos], [viagra] and other forms of highly spammy queries.
Forum discussion at WebmasterWorld, Twitter and Google+.
Update: It began rolling out 4:40pm EST:
@BtotheMcG it's rolling out now!
— Matt Cutts (@mattcutts) June 12, 2014
Comments:
Tian_Mian
06/12/2014 02:53 pm
No one writes here? Does everybody think they don't have spammy sites? Barry, looks like you got only white hats reading your stuff.
Barry Schwartz
06/12/2014 02:56 pm
What?
Ashley
06/12/2014 03:28 pm
About time Google updated Penguin to release those of us who have been devastated the negative seo.
Yo Mamma
06/12/2014 03:57 pm
Wha wha wha No changes to anything. I've given up expecting anything intelligent or cool coming out of G$$gle
Bryan
06/12/2014 04:25 pm
Hey Barry. Is Google asking you to call it the Spam Algo instead of the PayDay Loan algo? I thought Penguin was the web spam algorithm. Feels like I'm missing something here...
Barry Schwartz
06/12/2014 04:30 pm
Nope, Google doesn’t care what I call it.
Fedor
06/12/2014 04:59 pm
I miss the days when they didn't tell us anything, then at least you didn't have to worry about something until it happened. Now it's non-stop paranoia.
Jamo
06/12/2014 05:31 pm
The 'Chicken Little' effect. I saw a tweet from smx yesterday that said something to the effect of 'if you search the same keyword all day long, you'll see many different results' - why is that sooooo hard for people to realize? Stop obsessing over ranking changes that are < 8 spots it just gets Barry posting more speculation and its not helpful to the community at large. But it does make for good click bait!
wertwert
06/12/2014 05:36 pm
Actually I check rankings from the same internet connection, computer, time of day I see very little variance from day to day. Checking from different browsers and different speed internet connections does produce very different results. Largely due to google accounting for client side speed estimates on which results to return. It is a good reason to keep your page load times very fast. If I saw a lot of my rankings move 8 places in either direction overnight I'd !@#$ a brick. Maybe 1-2% of my pages google dance... overall it takes change on my part or a competitor's or change on Google's part (algo update) to see any change at all. Again this is because I keep all of the other variables unchanged when I measure.
James
06/12/2014 05:45 pm
Here he goes again ... wha wha wha
Jamo
06/12/2014 05:56 pm
Sure, that's fair. I'm talking about the inexperienced webmasters who have a page ranking #2 (and obsess over it), which probably has no business ranking #2 and it drops to #6 and they freak out and Barry posts about it. Did your traffic drop? Did you lose money? Ok, fine.
wertwert
06/12/2014 06:05 pm
Good clarification. Thanks!
Michael
06/12/2014 06:28 pm
Damn, thought you had more clout than that Barry.
Michael
06/12/2014 06:29 pm
He said, "No one writes here? Does everybody think they don't have spammy sites? Barry, looks like you got only white hats reading your stuff."
Yo Mamma
06/12/2014 06:30 pm
What is a TROLL? Here is a graphic of a TROLL. Its when every comment James makes is made after mine - Trolling my comment. He trolls BEHIND me. There is nothing flattering about being in the position closest to my ass. Whats the view like James? Attractive?
Fedor
06/12/2014 06:38 pm
LMAO This is hilarious. So you're saying that Google tests your internet speed before returning results? So it will return a different result if someone on your LAN isdownloading a movie and taking up most of the bandwidth than a minute later when the download is done? It certainly cannot do it based on provider because you can subscribe to a slow connection than your neighbor but they'd get different results because they have a faster connection? Same goes for mobile. Your mobile browser doesn't tell Google how many bars you have or if you're on 4G etc. Your speed can change at any time, especially if you're using wireless connections. This idea that your results are different based on internet speed is completely IDIOTIC.
Jamo
06/12/2014 06:42 pm
Matt Cutts cares about accuracy in reporting: http://searchengineland.com/googles-matt-cutts-metafilter-hit-algorithm-panda-penguin-193849
wertwert
06/12/2014 06:51 pm
https://developers.google.com/speed/docs/best-practices/rules_intro “Page Speed evaluates performance from the client point of view, typically measured as the page load time. This is the lapsed time between the moment a user requests a new page and the moment the page is fully rendered by the browser. The best practices cover many of the steps involved in page load time, including resolving DNS names, setting up TCP connections, transmitting HTTP requests, downloading resources, fetching resources from cache, parsing and executing scripts, and rendering objects on the page.” I commute on a ferry where the wi-fi is crazy slow. Same search is always shows consistent results for the connection. Only difference in the results is that slower/heavier sites rank lower on the slow connection. Test it for yourself. It also explains why browsers that use the same rendering engine see the same results but switching rendering engines gets you slightly different results. A few years ago everyone was in a stink that Firefox results were different... its just speed estimation quirks caused by rendering engine performance. Surprisingly Google actually wasn't singling out FireFox for special treatment...(Rare case when I defend Google) they were treated the same but produced different results. Test that too.
wertwert
06/12/2014 07:09 pm
Also (because the two go hand in hand) "Posted by Amit Singhal, Google Fellow and Matt Cutts, Principal Engineer, Google Search Quality Team" "Like us, our users place a lot of value in speed — that's why we've decided to take site speed into account in our search rankings." http://googlewebmastercentral.blogspot.com/2010/04/using-site-speed-in-web-search-ranking.html
Fedor
06/12/2014 07:32 pm
You're just accessing a different version of the index from a different datacentre. The delivery is likely based on your user-agent, caching and a random datacentre for load-balancing. It has nothing to do with the connection speed or rendering speed. You're talking silliness.
Fedor
06/12/2014 07:35 pm
The speed of your site. Not your connection speed. There's a big difference. Unless you're hosting your site from your mobile.
wertwert
06/12/2014 07:46 pm
That doesn't explain why it is always consistently different in the same way FOR YEARS or the fact the results change based on the browser rendering engine being used. aka Firefox getting different results. Find someone with a mobile hotspot and see for yourself.
wertwert
06/12/2014 07:47 pm
The two are hand in hand... if you know a client is on a slow connection showing the 15mb flash websites is a bad experience. If the connection is super fast page weight is less important as a factor.
wertwert
06/12/2014 07:58 pm
Looks like your mobile browser DOES tell you connection type AND it appears to be developed by Google. http://davidbcalhoun.com/2010/using-navigator-connection-android/ Visual indication of a device connection speed navigator.connection on Android 2.2+ switch(connection.type) { case connection.CELL_3G:
Fedor
06/12/2014 08:02 pm
Yeah, Sure, try user-agent spoofing and connection speed limiters/simulator. Do some real testing.
wertwert
06/12/2014 08:05 pm
user-agent spoofing doesn't do it... I've already DONE testing. You however are in the dark.
Fedor
06/12/2014 08:16 pm
You're still missing the point. Not all types are the same, just because someone on wifi doesn't mean it's any faster than 3g or 4g. There's no way to know how fast your speed is without conducting a test at each request because your connection speed can change in milliseconds. It is impractical to waste resources to analyze a connection that often, as a result it would be impractical for Google to provide serps based on your current connection speed. They can make guesses and that's about it.
Fedor
06/12/2014 08:18 pm
K, I'll live under my rock then.
Fedor
06/12/2014 09:16 pm
So I did some more digging into this. Some browsers can provide connection speeds through the Network Information API. The problem is that it's way too immature and there's no standardization yet. http://www.w3.org/TR/2012/WD-netinfo-api-20121129/ Quote: "One concern is that bandwidth may be hard to implement, can be quite power-consuming to keep up-to-date and its value might be unrelated to the actual connection quality that could be affected by the server." So even the authors of the API your articles references don't think it's practical to return updated connection speeds.
wertwert
06/12/2014 09:35 pm
I don't think Google needs this anyway was just pointing out it existed... by the time you load a google search page Google can know how you fair relative to your peers. Google has a ton of bandwidth so any difference in performance from one user to the next is likely going to be client side differences. Definitely with instant they have all the relative performance data they need to start making estimations. It would only take a "not good enough" switch to send a search to a bank of servers where "page weight" is a more important factor. I have seen evidence of this for a long time. The only thing I object to is you calling it IDIOTIC before actually investigating it. I don't mind being proven wrong.. God it is refreshing when I actually get awesome info that I don't have to prove for myself by myself. If it *actually matters* to you I sincerely hope you test it for yourself. At the end of the day nothing I say is going to make a lick of difference here.
Hex
06/12/2014 09:49 pm
Why noone ever kicked his butt for rolling out Panda, gay spam algorithms and shit?
Fedor
06/12/2014 11:39 pm
It's difficult enough to have an algorithm to rank sites but then sort them by amount of bandwidth available to a device? It just doesn't seem practical to me, just imagine the extra amount of computational power it would need. Google is for searching, it's the responsibility of the web developers to make a site work for all platforms and deliver the content efficiently to the visitors. Why would Google go out of their way to sort results by connection speed? What if a site is slow but provides the best information, why should it rank lower on a slow device? I don't see connection speeds or rendering as a major factor, especially when we're getting faster and faster speeds. Google is ranking how valuable the resource is, they shouldn't restrict access to it. But yeah, I guess I got too excited with the idiotic remark.
buxshots
06/12/2014 11:43 pm
Barry, do you have any info about a possible update regarding rich snippets (authority)? Maybe they did two updates in a go? -Because I got my snippets back this night. I hope thats a good sign...
Barry Schwartz
06/13/2014 12:15 pm
not seeing anything on that.
Barry Schwartz
06/13/2014 12:16 pm
wait until you see the video of the session, you will see how he said what he said.
Patti Paz
06/13/2014 01:04 pm
If Google begins to feels like it is not getting contact 24/7 attention, then it creates 'news' with non-sense post to where-ever . . . AND, it's biggest cheerleader normally passes the message. Let's all fall into line and give it the attention they are seeking.
Alex Garrido
06/13/2014 01:17 pm
I am seeing a lot of movement in rankings across several of my websites. Actually most of them seem to be positive. Nothing too definitive yet but it is not normal to see that much variation with multiple sites. Maybe the party just started? 8AM US Central
qqq
06/13/2014 04:41 pm
matt cutt version 22 will launch spam update N52 at next week. Reason for this update to remove all sites of people who not pay to google.
Ujjal
06/27/2014 03:25 pm
Hi Barry! I've got an odd problem. My job/resume related site was doing very well. I was getting more than $.40 CPC ($.40 - $.48) on average but suddenly it dropped on June 13. I'm now getting $.34 CPC on average. The most strange part is that the traffic is more or less same, only CPC has been dropped and hence RPM. I get 90% of my revenue from the US. I can not figure it out whether it is hit but PayDay 3.0 or FIFA World cup or Adsense Smart price. There was 2 big events on 13th June. PayDay 3.0 and FIFA world cup inauguration. If it was hit but PayDay 3.0 then traffic should be dropped, not the CPC. So it may be low bidding on job/resume related keywords during world cup or I am smart priced. I'm now waiting for world cup ending. Can you please shade some light on this problem?