| This 65 message thread spans 3 pages: < < 65 ( 1  3 ) > > || |
|Panda Metric : Google Usage of User Engagement Metrics|
User action is everything
| 8:09 pm on Apr 21, 2011 (gmt 0)|
I am surprised and amused at some in the community that point to a couple of individual metrics as the ones that caused xyz to happen in the most recent update. I think it is time to look at the greater picture of data that Google has available for analysis and interpretation.
Eric Enge had a presentation at PubCon Austin that he felt Panda pumped up to 20% user engagement metrics into the algo. It really got me thinking about the user engagement aspects indepth. In this socialized world, it just makes sense that Google would start using more engagement metrics such as demographic, psycho-graphic, and behavioral metrics. I started to put together a list of possible data sources Google could use as signals, and the list quickly grew large.
Most of the engagement metrics Google can use, will fall into the realm of user behavior. Those data sets can be combined with a successful search result into a powerful metric for your website. I believe that metric is now replacing Page Rank as the number one Google indicator of a quality site. I have been calling this mythical metric, the User Search Success Rate (USSR) or the Panda Metric (PM). This is the rate at which any search results in a happy searcher.
The metric starts before the user ever types in a query at Google:
1:Referral? How did the user come to Google? Was it from:
- a toolbar (Googles own toolbar, or a branded toolbar from a partner?)
- a partner site (AOL etc),
- a specific browser, Mobile, Desktop, Tablet or something else?
- a link on another site?
- a social association metrics? Did it come from a social site, and do we know who you are already? (Orkut, Twitter, Private Control panel such as wordpress?)
2: Location data
- IP address
- GPS Data available? Depends on device.
- Toolbar location data and history.
- WiFi network, Cell phone network or other ISP like location data.
3: Browser request headers
- Browser agent, platform and device data
- http accept: gzip, java, flash, etc.
- Screen size
- Toolbar metrics tell all (query string often included agent identifiers)
- Toolbar installation history and other history you may have already shared with Google. (such as version of toolbar)
4: Site Tracking and Advertising Tracking:
- What site did you come from and what did you do on that site? (if they were running Google Analytics or other Google Trackable metric)
- Both via Google Analytics and via Google site based advertising like AdSense or analytics (remember, you leak a referral every time you visit a page with Google code on it)
- Coming soon: +1 data from Googles' +1 service.
- My Google or Google Properties, Gmail, Youtube, etc.
- Sites you were logged into while viewing Google advertising from DoubleClick or AdSense. If you click through a login page on Wordpress at "foofoo.com" and then view an adsense at on that site - it is a good signal to track you by
At this point, Google knows who 70-75% (my guess) the users are doing any given query, and can guess accurately at another 15-25% based on browser/software/system profiles (even if your ip changes and you are not logged in, Google can match all the above metrics to a profile on you). That leaves less than 10% of the users in the world, that Google does NOT know. Of that 10%, they can later retro-analyze your profile again when you meet some criteria, such as logging into a Google service such as Gmail. (I'm not saying they care WHO you are specifically, just that you are User xyz that they can track over time)
Finally, after all that data, the user probably types in a query: (if the search didn't come from an off site query box or toolbar to start with).
- various psycho-graphical data. How the user types in queries into the auto fill box is indicative of users education level, often sex, and other psycho/demographic details.
- spelling, language, syntax, format, etc. All the variables of query can give clues as to who and what the users intent is all about.
- Mouse over preview data. What did the user do?
- Mouse tracking via the js mouse over data. (when you mouse over descriptions, that can be sent back to Google) Intent data.
- Multivariate testing metrics. As we have seen the last year, Google is SERP testing constantly.
Offsite Metrics: Finally the user clicks on a result and is taken to a site:
- AdSense or DoubleClick serve ads?
- Google Analytics running?
- What dose the user do while he is there? Click path.
- +1 buttons to come.
- How long does he stay on the page
- Does user visit other pages?
- Does user hit the back button and return to Google, or does he wander off to parts unknown?
- Toolbar data. Tracking and site blocking.
Unknown Crowd Sourced Metrics
- Google has been known to come up with unusual metrics and crowd sourced data. Take for example Google's Captcha system that is actually used to validate corrupted words from Googles book scanning project. I think it is a good assumption to think there is other data Google could mine that we are oblivious too. If you have played the game werewolf with Matt Cutts - you know he is a crafty one ;-)
After all that, we can quantify a metric (I call, The Panda Metric). It is an amalgamation of the above data inputs. This set of inputs would be relative to this query. They could also be weighted to relative queries (siblings, brothers/sisters, parents of the query tree from the root query).
How the Panda Metric would actually be applied only leads to more questions:
- would Google want people to leave a successful query and come back to Google.com? Or where would they want the user to go?
- Does a happy Google user keep using Google?
- What should Google do to retarget followup queries?
- Is personalization all it is cracked up to be?
- Does the Panda Metric result in higher Panda Metric scores?
- Is it a self-fulfilling or self-defeating metric that leads into a feedback loop - almost a race condition?
Any way you look at, that data when analyzed and applied within the algo could lead to a higher happy searcher result. I think the above data is partially what drove the Panda update. Why? Highly successful, high referral, low bounce, quality content, high engagement, and historical pages have seen a solid boost with panda.
| 3:53 pm on May 2, 2011 (gmt 0)|
> explicite feedback on SERP using "Block result from xyz site
I left it out intentionally. I feel that the signal is so small and noisey, that the value of it, must be limited. It's only value is in spotting stuff for hand checks. I don't believe it is going into the algo what-so-ever. Chrome has 5-10 million daily users. of those, there will be a pretty small percentage that have the extension, know what the extension is, and use the site block option on any kind of basis. It's main usage will be by competitors reporting each other. I bet there are less than 10k clicks on the 'block' button a day. The only place that is of value is in keeping the top searches clean via hand checks.
quiet the opposite and more of a nightmare really. I brought all this up, for the main reason of reminding people that what we think we know about the Google algo, is often 180 off. We can't assume we know anything without some empirical proof. That type of proof is next to impossible to work up in this day age. The data set is too deep.
| 4:19 pm on May 2, 2011 (gmt 0)|
OK, apologies for assigning motive where not intended (wishful).
I don't see any reason to fear what's coming though. Surely, true meritocracy on its way? I mean, it will become pointless gaming the algo to get to the top, only to be culled by user/ click data rapidly because you didn't add any value for visitors.
Means that SEO might become as much about making a worthwhile contribution to the web, as it is gaming google.
Hoorah the end of link begging/ buying/ swapping and creating!
| 7:51 pm on May 2, 2011 (gmt 0)|
Let's not forget that we talk about user engagement as a signal. It is NOT the salvation for ranking better per se. SEO has its classics, SEO has its link building and now SEO has just one more thing to keep in mind while working down that huge list of little things to do right...
| 8:24 pm on May 2, 2011 (gmt 0)|
I guess now when all is said and done we now can ditch PR (PageRank) and now use PM (Panda Metric).
| 9:01 pm on May 2, 2011 (gmt 0)|
From an article dated May 20, 2010:
Google showed an interesting chart yesterday at the first Google I/O keynote. The number of Google Chrome users more than doubled in less than a year from 30 million users in July 2009 to 70 million users today. It's worth pointing out that this is the number of active users, not the number of Google Chrome downloads.
Link: [googlesystem.blogspot.com ]
So about a year ago Chrome already had about 70 million regular users. I believe that this is enough for Google to collect meaningful data on user behavior. Also, contrary to what was implied in a previous post, Chrome allows Google to collect data for ANY WEBSITE that gets a significant number of visitors. Possible useful data would include:
1. A user bookmarks a page as a favorite.
2. A user saves a copy of a pge on their hard drive.
3. A user returns to the same page later.
4. A user visits other pages on the same site.
| 9:30 pm on May 2, 2011 (gmt 0)|
I don't think Google will ditch PageRank as that is what it got it to the top of the pile. Matt Cutts repeatedly states, that we should build our sites as if the search engines didn't exist and give our users the best experience and that is what will give us the best SERPs. I think that is what we need to focus on.
Up until now in the world of Google --Page Rank-- meant relevance. When I read the intro to this thread I thought that it was going to talk about how User Experience now equals relevance too. That seems to be the general consensus here, but I feel that Google will shy away from practices that expose it to anti-trust action. Overly relying on Data gathered from products like Chrome could start the ball rolling in that direction.
From my own experiences with the Panda update, from many sites in in different industries, one having PR 0 and two having PR6, my conclusion is that, while other factors may be in play, Panda has placed greater emphasis on on-site factors.
The evidence I'm looking at is a new site with PR0 in a fairly competitive niche, with only 3 incoming links, went from mostly page 3 results to page 1 and page 2 results after Panda. And a PR6 site with a very popular search term that has ranked number 1 in google for 8 years and finally fell to number 4. I had never optimized certain on-page factors because I believe, "if it ain't broke don't fix it". So I optimized for on-page, and it went back to number 1. (Are there any other success reports of a post-Panda revision?).
In this forum I see little attention to on-site factors. So I'm just wondering what people here see as the top 3-5 onsite factors?
| 9:46 pm on May 2, 2011 (gmt 0)|
|Matt Cutts repeatedly states, that we should build our sites as if the search engines didn't exist... |
So, paid links, web rings, reciprocal links, article submissions for links, directory submissions, forum link dropping, blog link dropping, reciprocal twitter follows and everything else they tell us not to do are really good ideas?
They're so contradictory sometimes, it's not even funny ... It's like none of them ever stopped to think about what people would [have to] do to get traffic from the web to their site(s) without search engines.
Personally, I wish they would make up our mind and get their advice straight.
| 10:03 pm on May 2, 2011 (gmt 0)|
Create the best "user experience" - and not to worry, you will rank high.
Yeah, sure . . . . . - have seen that dream before . . . .
| 11:28 pm on May 2, 2011 (gmt 0)|
|have seen that dream before |
Well it's worked fine for me these past 13years. How many of your sites rank number one for all relevant keywords, doubting the words of Matt Cutts?
| 12:03 am on May 3, 2011 (gmt 0)|
So I'm trying to understand... for a site that was hit in Panda, and does not have Google Analytics installed, would you recommend to install it?
| 1:44 am on May 3, 2011 (gmt 0)|
No, no Goog products, No WMT, No Analytics, No site search, don't make it easy for them to f*** with your site. Especially if you are white hat there is no reason they should f*** with a site.
You can get a site search for $5/month and analytics for $1/month, no need for their products. Don't leave footprints, they will track you down and treat you like a SE terrorist. This means they will give you a -50 penalty or will ban you from their SERPs.
I know they don't owe you nothing but don't be dumb, don't make it easy for them.
I think their site reviewers all over the world are underpaid and they will kick your site whenever possible, you know this is just frustration. This is my own experience.
| 1:52 am on May 3, 2011 (gmt 0)|
Brett, I think you overlooked Google Places as a factor.
For those of use running local directories you can now find a PR0 or PR1 site registered in Google Places easily outranking a PR6 page about the same location and this is completely new AFAIK, or in my niche experience.
| 2:42 am on May 3, 2011 (gmt 0)|
Agree suggy there is nothing to support this and being heavly involved I just don't see it. I see it as suggy suggested in the coming years but right now now it just isn't there. I know it is coming but in the present serps user data just doesn't carry that much wieght
| 7:39 am on May 3, 2011 (gmt 0)|
|Well it's worked fine for me these past 13years. |
Sure, if you niche is about lion toenail cutters, you wouldn't need links at all. Hell, you can do away with on-page SEO, but if you are in a competitive niche, you would be dreaming for a while.
| 9:50 am on May 3, 2011 (gmt 0)|
|Hell, you can do away with on-page SEO, but if you are in a competitive niche, you would be dreaming for a while. |
The PR0 site is for free online videos and it ranks top 9-30 for search terms such as :
watch category movies free
watch free category movies
Where category is a word like sci-fi or adventure. Maybe you would consider this a non-competitive category.
I still haven't figured out how that site can do so well with PR0. It has backlinks from my own sites PR5 and PR3 pages (that only have 30 and 3 backlinks going off the sites -- so it is a lot of link juice being sent). It is 100% optimized for onsite SEO and has no link exchanges.
Of my 5 sites, all have improved under Panda, with a few exceptions of terms going down a few notches (but I have a formula for a fix on those terms), all have few to none reciprocal links, 2 have PR6, 2 have PR3, 1 has PR0 (4 months old).
I also notice that you say on-page, where as I optimize on-site...huge difference
| 10:43 am on May 3, 2011 (gmt 0)|
|No, no Goog products, No WMT, No Analytics, No site search, don't make it easy for them to f*** with your site. Especially if you are white hat there is no reason they should f*** with a site. |
Wondering what you would say about Google Affiliate Network (as an advertiser, not as a publisher).
| 11:54 am on May 3, 2011 (gmt 0)|
|No, no Goog products, No WMT, No Analytics, No site search, don't make it easy for them to f*** with your site. Especially if you are white hat there is no reason they should f*** with a site. |
I am 1000% with SEOPTI on this one; when someone is interested in cutting in to your profit(oh, yes they are), why would you give them a tool(data) to accomplish just that?
Put it nicely: I don't trust Goog with my visitors Data, not even for a split nano-second.
| 1:29 pm on May 3, 2011 (gmt 0)|
|I am 1000% with SEOPTI on this one; when someone is interested in cutting in to your profit(oh, yes they are), why would you give them a tool(data) to accomplish just that? |
Don't want to get way off topic but I read that the "big river" site does this. They check their sellers to see what items really sells well and they will cut in.
| 1:33 pm on May 3, 2011 (gmt 0)|
I've been staring and staring at search results, both for my site and others, and trying to find some commonality. There's the usual on-page factors that have been discussed ad nauseum (thin content, dupe content, etc).
Brett's "user engagement metrics" makes 100% sense. Having said that, I don't see how turning off Analytics or Webmaster Tools would change user engagement or the on-page factors.
If turning either of them off would help, I'd do it in a split second. Does anyone think G can't measure all this without you having an account?
Brett, thank you for outlining your engagement theory. It's the best post I've seen from you since your 12 steps post years ago. (Or maybe you've posted better, but I haven't seen it ;) ).
| 3:12 pm on May 3, 2011 (gmt 0)|
Everybody is obviously working with different sets of information, knowledge and assumptions here, so without a baseline the discussion isn't very effective.
As far as Google products are concerned, they do, unquestionably affect search results. That means, in some cases, what you see, is different from what the guy sitting right next to you sees. You can test this, get a google account (mail, analytics, whatever), after you've had this account for awhile, log in and do some searches while you are logged in. If you note your results. Log off, launch another browser that you rarely use or that has all cookies removed and search again (or just close your browser, delete private info like cookies and relaunch). You will see different results (at least in very competitive categories). Previous searches also influence the adwords you will see (as long as cookies are not deleted). If I have the toolbar installed and am not logged into an account, I still see different results than if I am logged into an account.
Does geolocation affect search results? You bet, get a UK vpn and search. You will get different results than if you are located in the USA (e.g. have a US IP address). This is again, unquestionable. Google country-specific sites also give different results for the same user location. To test, search for something in Google.co.uk, or google.au and watch the results change.
Sure, that is Google improving our user experience. But this is not the same question as saying, what is the baseline search result. In other words, my objective is to have the best SERPs possible for the vanilla search (i.e. the user who has no google search history in his cookies, etc) in whatever location he chooses to be. It is not possible to control what Google decides is a better result based on what the guy looked for over the last week or where he is located (although you can optimize for country-specific sites - or for Places, which is outside the scope of this discussion).
This means as far as SEO is concerned, if you wish to set on-site factors aside, a user experience can only be based on the way the user interacts with the site (page views, time on site, frequency of return, etc), assuming google actually has the capacity to measure such factors.
| 4:58 pm on May 3, 2011 (gmt 0)|
Google knows two things: bounce and time to adsense click (TTAC)!
I think going forward, bounce, blacklisting, TTAC will become more important.
This is the best feedback to google on whether offering a page to users is likely to be worthwhile or not. This may ultimately sink pointless/ waste of click sites. In the future, it won't be as simple as getting the SERP position; you'll need to keep earning your place there too!
| 5:15 pm on May 3, 2011 (gmt 0)|
@sundaridevi : correct observations.
Similar metrics, based upon data collected by Google in various ways, are also supposed to be used for AdSense in order to optimise revenue -- but AdSense relevance is abominable most of the time - thus the low CPCs most publishers experience.
Theoretically, it makes sense to use such metrics in influencing search results as part of an algorithm prevalent in on-page-content RELEVANCE factors. However, for the time being, there seems to be a considerable gap between theory and practice.
In the meantime, of course, Google has an army of low-paid idiots manually penalising sites, or group of sites, according to their "judgment", psychological problems and their own "interpretation" of what Google has provided them as guidelines . . . .
| 11:46 am on May 4, 2011 (gmt 0)|
|Theoretically, it makes sense to use such metrics in influencing search results as part of an algorithm prevalent in on-page-content RELEVANCE factors. However, for the time being, there seems to be a considerable gap between theory and practice. |
Yes, but it seems plausible that Google today, places more emphasis on a site-based view of the web, as opposed to a page-based view of the web as it did in the past.
What I mean by that is that in the past, a page had PageRank and apart from what was on that page, PR was the most important determinant of the SERPs for the page. But today I think that the relationship of a page with other pages on the site and the influence of what is on the other pages has become more important. Things have been heading this way for a while (i.e. the talk here of a -50 penalty, etc, is evidence of a site based view of the web).
| 4:58 pm on May 4, 2011 (gmt 0)|
>site-based view of the web, as opposed to a page-based view
Oh yeah. I think everyone agrees there is an 'authority' metric (aka: SiteRank/) that is sort of a cumulative page rank for the entire site. The theory has made the rounds under several different names since the Florida update in 2004. (brand authority, doamin rank, domain authority, authority rank, Site Rank...etc)
| 6:20 am on May 5, 2011 (gmt 0)|
>site-based view of the web, as opposed to a page-based view
The way I see it...
Page dropped a few places (maybe up to 10) on competitive phrase = mainly post-Panda site authority recalculation pulling down
Page dropped off the graph = page has failed document classification part of Panda algo; Google sees as potential spam.
Anyone beg to differ?
| 9:11 am on May 5, 2011 (gmt 0)|
Didn't somebody say there are 200 factors? I don't think it's possible to micro-analyze it. What i've said is trying to paint a general picture in broad strokes.
But I'm still surprised that nobody has offered to mention any on-site factors that they think matter...other than duplicate content and "thin content". The so-called "thin content" has long been an SEO factor measured by other easily measureable quantities, some of which the majority of SEO experts will tell you don't matter at all.
| 10:15 am on May 5, 2011 (gmt 0)|
I was talking at a macro level!
Obviously there are many ingredients to the site authority rank, which have clearly changed under Panda.
Likewise, google has a set of parameters for determining content is unworthy (shall we say) and not all are on-page factors. I think I have figured out the gist of thoese.
On a page by page basis, Panda seems to work at a search phrase level. A page that no longer ranks for its primary search term under Panda, still ranks for secondary terms on the page (ie. tanked for Blue Widget; riding high for Blue Widget for Woman)
| 9:05 pm on May 5, 2011 (gmt 0)|
That is a facinating look into the theoretical mind of Google BT, but with all due respect, I don't see how all those metrics really boil down to how Google currently sees user intent. Its just a tad over doing it on the analytical side, don't you think?
I'm sure it is amusing to think that Google can conjure up such an elaborate way of algorithmically serving results using all those metrics, but looking at the SERPs before and after the Panda update, it seems to me that it really does boil down to a handful of factors. I still think links (irrelevant or not) are the major factor, but i believe those were simply a threshold for the ones who were to lose versus gain in a "content fight". So for instance, if you had lots of links, with the same content as another, you would most likely gain rather than lose. There is still tons of crap sites ranking right now, and if the quality of one's site reflected through user behavior was really a metric Google used, I highly doubt the SERPS would look as they do now. At least from my viewpoint.
| 4:54 pm on May 15, 2011 (gmt 0)|
I've just been reading Bill Slawksi's article about Microsoft's Approach to User Feedback Data [seobythesea.com], and it struck me that all these metrics are potentially available to Google as well:
|Some user behavior signals tracked may involve navigation through a site or the display of a page, such as when: |
A hyper link has been clicked to navigate to a different page
The history is used for navigation to a different page
The address bar is used to navigate to a different page
The favorites list is used to navigate to a different page
A document has been completely loaded and initialized
Scrolling is taking place
A document is printed
A document is added to the favorites list
The window gains focus
The window loses focus
A window has been closed
The user selects, cuts, or pastes portions of the displayed page
Navigation back to the search results page
Other signals could involve:
User dwell time on a page
A new query initiated by the same user
Other sequences of user behaviors
Other information about searchers' behaviors are likely to be collected such as when a hyperlink is clicked, the position of the link may be recorded, the size of the content involving that element (image, anchor text length, area of content where the link is located, and the type of content may be identified as well.
| 8:56 pm on May 15, 2011 (gmt 0)|
Bill also did an analysis of Google patents that shows some of the stuff they have the ability to look at (and how they could potentially use it), like:
copying and pasting (site engagement metric)
bookmarking (site engagement metric)
time on site (site engagement metric)
links clicked (site engagement metric)
mouse-hovers in SERPS
lots of user-based probabilities in ordering results
| 8:00 am on May 16, 2011 (gmt 0)|
@ Sundaridevi --
That means we can make all flash websites (as Google does not exist) and they would reward you with traffic?
| This 65 message thread spans 3 pages: < < 65 ( 1  3 ) > > |