|Google Updates and SERP Changes - December 2010|
< continued from [webmasterworld.com...] >
I know tedsters has stressed this but what we see ourselves in the SERPS doesn't always help. We are getting pummeled by traffic on one keyword that is relevant. I can't find it and neither can any friends buts it's converting at a great rate. When I perform that search i get no results. Oh the games....
I'd say 20% is a close number compared to two weeks ago. Google traffic is off 35% over these days last year. Yet business is strong, and the bounce rate is low.
The shopping SERPS are impacting more this year because exact matches are turned down. Theres s bunch of products that we were 1-3 on last year where we are gone and so is everyone else, replaced by unrelated stuff...think:
Product whiplash from Appalachian co
Man falls from tree in appalachians suffering whiplash
Walking the Appalachian trail
Blah bla x 7
The exact match of the product name yields no results in holiday search 2011.
I was laughing tonight when I saw a site on amazon that sells helmets also is selling tools. They rank near the top on many items in google. They have no ratings and appear to just dump feeds. #1 in google, ahead of the manufacturer and a world renowned climbers site. That's the lunacy in all this, amazon isn't an authority on rock climbing helmets, crucial memory, or even sweat socks sold by 3rd parties.
This is as bad as the earlier wiki is 1st which was adjusted out. An unknown third party selling on amazon using content that's identical on 5k websites prior...doesn't make amazon the expert.
Im betting a lot of lost traffic went to amazon. They were up 25% yesterday, a lot of us were down 25%. Thanks google im sure amazon needed that boost.
[edited by: tedster at 6:51 pm (utc) on Dec 1, 2010]
IMHO, it would be unfair and unwise to group websites as a neighborhood. This would be an open invitation for squatters (scrapers) who want to borrow some of your content to be included your niche neighborhood. We already experience that. I love competition, but when a scraper or aggregator comes along, grabs up some of our content and optimizes an MFA site in my neighborhood, that brings property values down to slum level. Just because someone is stealing content an slaps a site up in our niche, doesn't make them relevant. I still believe branding & trust should be primary ranking factors . Branding takes a monetary commitment on the part of the business owner & trust takes time to earn. I seems to me that everything else can be "gamed".
If they want to do a neighborhood thing, create a virtual, visual "main street" and sell or rent the "real estate" as if it's a brick & mortar. That's where this is probably headed anyway, and in some ways already is (Adwords).
I used a proxy to search Google this morning and the results were very different than I was used to seeing. (I use Firefox for general browsing, and block as many google related cookies as I can to avoid being monitored) Instead of seeing 2-4 variations from the first search result, that same thing was repeated throughout the top 25 or more.
Essentially, 6-7 websites were split among the top 25 search rankings, when using an anonymous search.
1-3, Site 1
4-7, site 2
8-10, site 3
11-12, site 4
13, site 5
14-15, site 6
16-19, site 7
20, site 8
21-24, site 9 (which is the SAME as site 7, with a hyphen in the domain)
I don't recall seeing results like this before with so MANY duplicates from similar domains. The same trend continued throughout the top 100 results, with sites 1-9 often repeated throughout as well.
The word 'review" was a part of my 3 word query.
Just for kicks, I did the same query using Chrome, logged in to Google services. The "Site" portion, is the same site as the proxy search above. HUGE differences when logged in, and not one single "block" from the same domain.
1 - Site 7
2 - Site 2
3 - Site 1
4 - Site 2 (again)
5 - Site 9 (same as 1, hyphen in url)
6 - New site
7-9, all new sites
10 - A Site I am associated with.
Very odd results for me this morning, but also very telling that anonymous browsing just took a hit.
I've too have been toying around with anonymous proxy searches and the results have also been quite bad, lotsa borderline spam pages and duplicates. Much worse than if I was logged in and using chrome. It would seem that personalized search is getting better but anonymous search is still in the dumps.
Strange thing that I've noticed is that searches seem to come up better using chrome compared to using firefox or IE.
@ryanep - [edit, removed some of my own off-topic]
In this case, it seems you are clearly saying:
- Don't search behind a proxy for better results.
- Stay logged in for better search results.
- Use chrome for better search results.
This would suggest that any kind of neutral search results from Google is completely out the door, and the ONLY way to get search neutrality, is to use a different search engine, and a different browser.
Is that what you are finding also?
@mhansen - Yes, that's what I've noticed over the last couple of days.
I know these observations when put together may sound very "conspiracy theory" but from what evidence I've gathered and results from the different search parameters and sessions I find it to be true.
My theory is that when you're logged-in, G has a much better idea of what you're likely to want and thus has more factors to use when deciding which results to serve. The higher number of factors would explain the better results.
If you're completely anonymous the number of factors available to G are less. This in itself may already explain the poorer results.
Combine this with the previous proposition that G's "trust engine" may be broken or that any other part of their search components aren't working 100%. Now we have a situation where anonymous search will rely more on these broken components (compared to personalized search) just because other data isn't available. In theory this would make anonymous search even worse which is what I'm seeing on my end.
@ryanep - OK - Now I'm starting to get it! I can also see where a conspiracy theorist may have a field day with this one! Almost like Google "Broke Itself", to force people to login or allow tracking if they want relevant search results! (or at least what Google feels is relevant for that specific user, based on profiling)
Purely presumptuous, but I wonder if we can expect a statement saying something like: "We have become so complex and accurate, that the ONLY way to get relevant search results, is to sign-up free, and agree to let us to personalize Google just for you!" They could even follow it with a cute little pre-cog snippet like: "We Already Know You Want To".
We may have a choice to make in the new year then... get personal with Google, or find a different way of getting relevant search answers.
I don't allow cookies etc and use the following:
Plus I have "Google Global" installed which gives 5 instant result comparisons between US, Canada, UK, Ireland and Australia.
It's interesting to note that the two .coms give different results for me!
[edited by: tedster at 10:02 pm (utc) on Dec 25, 2010]
[edit reason] made the complete url visible [/edit]
I used to see 3000 instances of a top converting keyword in search referrals per month(I know it is a converting keyword) and now I am seeing about 500 instances. I get certain impressions -
1. Google may have reserved the top converting traffic for it's advertisers.
2. Google may be hinting it's time for us to move to Adsense. It's like "it is in your own good that you shift to Adsense".
And I am afraid, this situation does not seem to be middle of the road construction work where they are experimenting and tweaking. I guess this is permanent.
Well, I was kind of expecting this. Although it's a big move and very shocking to most of the webmasters, it was logical for a company as big as Google to take these kinds of decisions.
They have solved many problems with this algo change and that's the brilliance of their engineers behind a great company. Although I m pi^^ed off, I can't help but appreciate their brilliance.
1). People were saying Adwords is not performing well. The bid prices have gone up, conversions aren't phenomenal as compared to what you could do with organic traffic and so on.
2). People prefer affiliate to adsense. In this process they make thin affiliate sites often times poor quality sites that Google is not very proud to display in serps.
3). People siphoning off long tail traffic and spam the serps in long tail.
Current scenario seems to take care of all 3 problems. It's like -
1). You want to earn good money - GO ahead and build a quality site with immense user value. Gain trust and have an awesome retention rate. And USE ADSENSE. We'll make sure your income is good.
2). Want to go for affiliate money. We have the best converting traffic in the world. Sign up for Adwords and let's start.
It's like Everyone else, keep on wondering what happened to my site.
One more thing - even if you see spam sites with lots of #*$! backlinks ranking in serps, that does not mean they are also getting a lot of traffic out of Google. That's what traffic shaping and normalization is all about. I sent 100,000 crappy forum profile links to one of my site, went to #3 for a very competitive keyword but the traffic increase was only 40 more visitors per day. :)
Take it this way, if WH are finding a tough time to capture good traffic out of google, BH can't do much as well.
|2). Want to go for affiliate money. We have the best converting traffic in the world. Sign up for Adwords and let's start. |
You must not promote affiliate sites with adwords other wise you would know your statement is totally without merit. It's not a matter of if they will the throw you out, it's a matter of when.
Google has been actively throwing out affiliate site accounts from adwords whether they are thin or not. They do not want affiliates advertising with adwords.
|Google may have reserved the top converting traffic for it's advertisers. |
Webmasters have been suspecting this since Google began their advertising programs. I still see no statistical evidence for it, and I've worked with a lot of sites - all flavors. That said, advertising done well can definitely bring revenue, as can running Adsense. But this forum is here to talk about organic rankings, and organic traffic is still available, even though there are a lot more players trying to carve a piece of that pie.
It takes more rigorous and detailed analysis these days than it used to. Generalized ideas and "braod brush" ideas do not usually bring the clarity needed to compete for organic visitors.
Just as an anecdote, I work with one site that is now 11 years old and has never paid for Google ads, although we do advertise on other sites in the industry. Rankings have never been better, including two "trophy terms" where we've been one of the top three for 8 years.
We're currently at #1 for the biggest money maker (just climbed back up from #2 this week) and #3 for the other - the second biggest. The past month or so was there a bit of wobbling, and now a different internal page ranks for each of the two big terms - the home page used to rank in each case.
|...and now a different internal page ranks for each of the two big terms - the home page used to rank in each case. |
I have noticed too that Google sometimes ranks home page rather than originally intended page and also some changes in replacing home page with internal page in SERPs in the last month or so.
I have a question. Lets say you have a site about colorful widgets, and three main categories: Green widgets, Red widgets and Yellow widgets. You optimise and intend to rank each of category pages for respective widgets, but Google decides to rank home page for all three keywords instead - what do you do?
Do you try to "force" category pages to rank or, seeing that Google is ranking home page well, do you change the strategy and support the home page to keep ranking for all three (and why)?
I could understand forcing categories to rank if the home page is not ranking well enough. I could also understand trying to force category pages to rank if the site is about more than just about colorful widgets (e.g. colorful widgets and different sized gizmos in roughly equal proportion - where these in this example these would be more unrelated terms), so if the home page ranks for each of colorful widget category, but not (or not well enough) for gizmos, this could mean that Google may not be understanding (for whatever reason) that gizmos are equally important too.
But if Green, Red and Yellow widgets are close enough (lets say, they can be directly connected in wonderwheel) would you support home page ranking for all three if Google is already seeing it like this, despite having distinct category page for each?
Or would you do your best to replace home page with category page in SERPs for each?
[ADWORDS is the advertising program. ADSENSE is the publishing program. There's a difference.]
|but Google decides to rank home page for all three keywords instead - what do you do? |
You thank your lucky stars that Google ranked you at all! ;^)
|We're currently at #1 for the biggest money maker (just climbed back up from #2 this week) and #3 for the other - the second biggest. The past month or so was there a bit of wobbling, and now a different internal page ranks for each of the two big terms - the home page used to rank in each case. |
tedster, can you give us some tips? How does your site differentiate from an average run of the mill seo'd site? Or What are the 3-4 most important factors you believe that have helped your site achieve this?
Would you give number 1 spot to Links or Content or User experience (retention) or Branding? I am guessing your site have all 4 good but how will you arrange them in descending order of preference? Have you got some very high quality or trusted links?
How much effort and how much return? :)
Also, I am curious whether you have seen the following on this site of yours -
1). Decrease in traffic even though maintaining the top ranking.
2). Drop in conversions. I have seen traffic coming more from Japan, Pakistan, Indonesia increased and thereby drop in sales.
This is the same site that I was discussing in the thread What is the smallest thing you have done that has made a ranking difference? [webmasterworld.com] - the one where making sure the vocabulary used showed a natural variety of co-occurring phrases.
The biggest factor for this site has been its content and focus on building a regular user community. We constantly publish new content in text, audio, video, news - and we publicize it through an email subscription list. Over time, the site has attracted a lot of backlinks for its best content - and also built strong partnerships with other businesses in the same niche.
The focus has been on building that subscriber list by offering "must have" content. The actual conversions happen over a longer cycle... the main offerings are not impulse purchases, given their 4 to 5 figure price tags. So over time we have added a lot of related but low cost products and services. That helps visitors experience the company's service and it build trust for the big sales.
Yes, we have seen some of the traffic anomalies you mention, and a lot of short term wobbles. As Google evolves and fine tunes their services I expect to see more. That's one reason why we strengthened our focus on building a community of regular users and buyers.
Today, Google traffic is as high as it's ever been - over thousands of keywords. From the low bounce rate and improving sales, it also looks really appropriate for the business. That could change tomorrow, and we know it. The "website" now represents a solid business and it has become a recognized brand. That's where we focus.
|Today, Google traffic is as high as it's ever been - over thousands of keywords. From the low bounce rate and improving sales, |
So far today I'd have to concur, but December 26th has historically been a huge spike day for us. It's starting out slower than usually, but no complaints on traffic sales conversions so far today.
Here's the rub though...
Our SERP's have not changed at all in the past few weeks...but traffic and sales are suddenly up, so this again points so some sort of "evil, anti-competitive" traffic shaping program. ;^D
Huge opening for programmers to develop "virtual users" using proxies and loading up "test" info to find SERPS with different parameters and "user settings".
An extension of that would be to again use proxies to "shape" web page profiles that the webmaster is seeking. 10000 interval visits a day by a proxied bot performing what the webmaster sets it to do is very easy.
I see this traffic shaping exercise to be totally open to spam and manipulation. I honestly hope big G has a big ACE up it's sleeve that we don't know about yet.
I'd like to call upon adventurous programmers to try develop this bot and testing it to see how it can "shape" SERPS. It would seem that there is increasing evidence from many fronts that traffic shaping is well and truly here to stay. Let's see what we can do with it.
If we do understand this aspect of Google correctly (and that's certainly not a sure thing at this point) then there is a lot to discuss along those lines. Let's keep that discussion in the main thread - Google & Traffic Shaping [webmasterworld.com].
That way we can keep this current thread focused on the algo changes and updates as we notice them.
@ Tedster -
Could you possibly add a bit of qualification to the current results you are experiencing... I think it would help to understand the budget or general size of the team that works on the site?
For example, if you're talking about the results of a $50,000/month budgeted team of developers, content producers, graphic designers and marketing professionals, its going to take on a very different view for the guy/gal who manages a 2-3 person e-commerce business, shipping from a small warehouse, and turning $35k/month in total revenue.
This business has three full time staff and four part time - plus me as the only person who has touched the website directly since the year 2000 (and it does look a bit outdated at this point.) A small amount of graphic work is outsourced, and a lot of web content is created by "celebrities" in the field. They contribute for free, just for the visibility they get from being featured on the site. Most of the sites I work with are major enterprises, but not this one.
I don't even know the budget picture, or how it breaks down between offline and online. But the online part is a lot less than half of the $50,000 a month that you mentioned. A secondary part of the business requires some inventory and leasing of a small warehouse facility with one part-time shipper. The core business requires no inventory at all.
The almost ironic factor here is that early on we realized the importance of not depending on search traffic as the only source of revenue. As soon as all our revenue streams became solid, Google rankings also stabilized and then began to grow.
And this month, after more than 11 years in business online, the search traffic is the best it's ever been. And we're making solid use of it while it's there - getting newsletter sign-ups and many other kinds of customer retention activities. We know that it could go **poof!** in January.
GREAT info Ted - Thanks for the deeper insight!
Two sites in the same sector - one is a retail site and one is a review site.
As of Dec 25th rankings for the retail site have started to dwindle and at the same time Google is sending a lot more traffic to the review site - I'm assuming off the back of improved rankings.
Is there any chance this could be caused by a seasonal algorithm change being turned off?
Note - I haven't been working on the review site at all for the past few months so the increase in traffic came out of nowhere.
Huge rebound started December 26th and ended this morning...welcome to the wacky world of Google!
backdraft, when were your site(s) affected? Did you make any changes to the site? What kind of rankings did you lose?
@brinked - Made basically no changes to site. Traffic and sales just exploded on December 26th. This is a typical pattern I see every year the day after Christmas. It usually stay very hot until early April...last year it stayed good until May 18th. So far one sale at 7 am. It's nearly 3 pm and still nothing else. It's as if G decides to cut you right off at the knees. They can just as quickly stand you back up. I've never seen these yo yo patterns more than in 2010. Spot checking a few dozen varied key words and phrases shows no decline in SERP for my site. Very Odd...zombies are back?
Seems like the rankings you are seeing cannot be what the average user is seeing then, right? Either that or there is a crazy drop in search volume all at once over lot of keywords - and that is not going to happen.
@tedster - I wish it were that simple. I have colleagues in Florida, New York, Arizona, Texas, Illinois, Pennsylvania and Michigan - all report similar results for about a dozen of my top key words, phrases.
My site hasn't changed, but traffic is still turning on and off with an almost predictable regularity. It's almost as if when Analytics detects a rise in my traffic, they counter by slowing it down to a "normalized" amount. My traffic last year went through the roof and stayed there for three months. Same pattern in all previous years. If there really is traffic throttling, I have no clue how they are doing it when I'm right up there in the SERPs.
Interestingly - My competitor seems to be on a see-saw with my site.
Since the competitor downloaded and reverse engineered my site, I know how to access his sales figures.
When traffic is running the "normal" pattern, as it has for years, there is about a 10 or 20 to 1 sales ratio to my competitor, who by the way, ranks well below me and get's less than half the traffic I get...of course he copied my traffic tracking too, so I have a direct comparison which comes in handy.
When Google seems to decide I've had enough traffic (as reported by GA), my traffic drops and the sales ratio is greatly reduced, sometimes even reverses, but just for a few hours or a day max. His traffic rises.
The competitor has stolen a lot of my sales copy, so I sometimes wonder if G sees the two sites as very similar, which they are due to the competitors theft. Problem is, he changed things just enough to avoid the hassles of a lawsuit or DMCA action.
Have you checked your DNS thoroughly?
If your rankings don't change much but traffic does change dramatically, then it doesn't sound like a Google mechanism to me. DNS cache poisoning is a very black hat technique that can siphon off search traffic, and the sneakiest type doesn't steal all of it because you would be in an alarmed state way too fast. On again, off again throughout the day would be the pattern that gets used instead.
Along those lines, I always check DNS for clients, and it's very rare to find someone with no vulnerability. See this thread for more [webmasterworld.com...]
[edited by: tedster at 3:41 am (utc) on Dec 30, 2010]
@tedster - (IP - Intellectual Property) My initial concern is that Google might think the two sites are dupes, since this joker stole so much sales copy.
Thanks for the DNS tip though...at least that gives me one more avenue to explore. I'm sure we're all feeling like Indiana Jones searching for the holy grail of SERP understanding.
It's very much worth checking out. The fact that you and your SERP competition are on a traffic see-saw also smells very suspicious to me, and that's not a Google suspicion.
@tedster & @backdraft7 It looks like backdraft is describing some parts which are the same as I posted here : [webmasterworld.com...]
My traffic is also compensated once I get more then google wants me to have.