| This 77 message thread spans 3 pages: < < 77 ( 1 2  ) || |
|First it was Position 6 - Now Position 4 Gets Strange|
| 1:08 pm on May 5, 2008 (gmt 0)|
The result in position 4 has been rotating between three different firms all day for our industry's top value keyphrase. All other listings above and below remain constant.
| 10:26 pm on May 20, 2008 (gmt 0)|
I've been working to stabilize a site that keeps cycling from the second page, up to position #4 and then slding back down to position 12 - often in one step but sometimes in several steps. The changes are very rapid, often within one day, By the way, this is a competitive and highly monetized phrase.
One factor I started studying is the size of the competition websites. These results seem to show that size of the domain may be a ranking factor. Also, before the cycling began, the site was very stably in position #4.
|URLS INDEXED |
Position 01 -- 1,570 Position 12 -- 1,040
Position 02 -- 12,100
Position 03 -- 99,800
Position 04 -- 87,000
Position 05 -- 27,700
Position 06 -- 22,100
Position 07 -- 242,000
Position 08 -- 1,630
Position 09 -- 124
Position 10 -- 689
Position 11 -- 125
Position 13 -- 976
Position 14 -- 13,600
Position 15 -- 401
Position 16 -- 1,660
Position 17 -- 10,400
Position 18 -- 1,490
Position 19 -- 3,750
Position 20 -- 686
|AVERAGE URLS INDEXED |
Average (#01-05) -- 45,634
Average (#05-10) -- 53,309
Average (#11-15) -- 3,228
Average (#16-20) -- 3,597
The anomaly in position #1 is probably due to having backlinks that go way past any of the other sites,
So what do you think? Is size of site a potentially big ranking factor? Then other factors would help Google decide which of the smaller sites deserve a shot at position #4 (or wherever they are currently testing.)
Also note - in the time it took me to write this post, position #4 went from being a website to being News Results (Universal Search). Some times in the past, News Results have appeared at #10 but usually they do not appear at all.
| 11:12 pm on May 20, 2008 (gmt 0)|
As exciting as it is to find every single one of the 400+ "tweaks" that affect rankings, I equate much of Google's algo to determining the weather.
When it's all said and done, the weather depends on 2 things. The sun and the earth revolving around.
I'll let you figure out what parts of the "algo" equal the sun and earth...
The FIRST thing i would be concluding from your data is NOT my sites pages per se but the ENTIRE SUN PATTERN that is only reflected in your neck of the world.
(for those who don't like analogies)
|The anomaly in position #1 is probably due to having backlinks that go way past any of the other sites, |
Hmm, almost like Goog is counting ALL of the #1(1-3) sites' backlinks all the time and NOT counting and THEN counting the pages(and links) of the sites that keep jumping around.
Anyone who's creating new content should be noticing the exact same phenomena with their "link-bait" pages.
(see earlier posts about what it is NOT)
| 2:22 am on May 21, 2008 (gmt 0)|
|So what do you think? Is size of site a potentially big ranking factor? |
Absolutely I would agree Ted. Even more important, do a supplemental test on those websites and you might find the same curve.
A website I watch is flickering between first page positions and 11-13 on page two. Recently there were some higher pagerank quality links secured to the site.
On the DC's that correlate with better positions, the number of pages in the site, and most importantly, the number of detectable pages not in supplemental index are both up considerably.
| 2:57 am on May 21, 2008 (gmt 0)|
|the ENTIRE SUN PATTERN that is only reflected in your neck of the world. |
And thus we have the value of forum discussions. Has anyone else examined the SERPs in their neck of the world and seen something similar?
|do a supplemental test on those websites and you might find the same curve |
That idea popped into my mind, too. I'm going to do it for this search at least, and also run the "indexed pages" analysis for some completely unrelated searches and sites. I'll report back, whether the results reinforce or work against my idea.
The reason I am particularly interested in this one, is that the SERP is showing intense churm - the topic of this particular thread. I haven't found another SERP quite like it in my collection so far, and it seems clear that different search areas can be treated quite differently.
| 6:35 am on May 21, 2008 (gmt 0)|
Tedster, how can you stabilize anything in such an unstable situation?
Do the sites with more pages also have more backlinks from external sources?
Do the large sites do a lot of internal linking or does it seem unrelated to this?
Does it seem like sometimes size matters then when the order changes that size no longer affects the results?
Are the pages related or are they like Wikipedia with many topics?
Inquiring minds want to know. ;)
| 8:57 am on May 21, 2008 (gmt 0)|
This makes me wonder if there is a tipping point related to the volume, position on page and anchor text of internal linking.
It was reported on the changes thread (old age prevents me from remembering by whom) that the writer had done some research which implied that internal linking in the menu, header and footer had been devalued. Pure speculation but what if, on evaluating the results of this, the folks at Google have spotted that it has had some unfortunate effects and large, authoritative sites that were ate the top of the rankings because of their internal linking dropped below where they would want them to be.
Having spotted that bug, if there are other effects, that fit the main objective of the change, then they have to do try something else to get those big authority sites back to where they deserve to be. I wonder if what we are observing is a knock on of that.
| 1:07 pm on May 21, 2008 (gmt 0)|
An interesting update to these changes.
Today for the first time we saw #3 shuffled to #7 for a short period. Up until now we hadn't seen any shuffling for results 1-3.
|Pass the Dutchie|
| 1:49 pm on May 21, 2008 (gmt 0)|
I don't suppose your figures include pages on a subdomain? I would say that size does matter.
| 4:19 pm on May 21, 2008 (gmt 0)|
if the size of the site helps place the site in the serps, why is 14 and 17 so low? and why is 9 so high?
| 5:05 pm on May 21, 2008 (gmt 0)|
>> Do the sites with more pages also have more backlinks from external sources? <<
There's a wide variation here in terms of backlink quality. Most of the sites with large numbers of links reported have done huge and very spammy link programs. Here's the backlink numbers from Yahoo. All these ranking urls are home pages, no internals.
|BACKLINKS from Y! |
Position 01 -- 205,000 Position 12 -- 1,140
Position 02 -- 140,000
Position 03 -- 149,000
Position 04 -- 973
Position 05 -- 15,500
Position 06 -- 15,700
Position 07 -- 33,200
Position 08 -- 33,700
Position 09 -- 2,700
Position 10 -- 32,500
Position 11 -- 24,200
Position 13 -- 15,700
Position 14 -- 315
Position 15 -- 102
Position 16 -- 6,840
Position 17 -- 40,500
Position 18 -- 8,190
Position 19 -- 15,600
Position 20 -- 1,340
>> Do the large sites do a lot of internal linking or does it seem unrelated to this? <<
The smaller sites are US only, and the larger sites are internationalized - they generate a lot of pages dynamically by plugging in city and country name variables. internal linking seems weak, but form spidering may be quite strong in these cases.
>> Does it seem like sometimes size matters then when the order changes that size no longer affects the results? <<
Of the top positions, only #4 seems to be unrelated to size of the site when things shift. Also, as of today, the site that replaced my client at #4 has now gone to #3. this is the site with 973 total indexed pages. This site seems to be sharing in the experimental churn through #4 with my client's site, but today they went up.
Also as promised, here are the numbers for pages in the regular index. I don't see much pattern here.
|REGULAR INDEX URLS |
Position 01 -- 840 Position 12 -- 193
Position 02 -- 8
Position 03 -- 1190
Position 04 -- 746
Position 05 -- 42
Position 06 -- 143
Position 07 -- 551
Position 08 -- 54
Position 09 -- 13
Position 10 -- 33
Position 11 -- 9
Position 13 -- 104
Position 14 -- 525
Position 15 -- 55
Position 16 -- 69
Position 17 -- 187
Position 18 -- 307
Position 19 -- 1
Position 20 -- 226
>> if the size of the site helps place the site in the serps, why is 14 and 17 so low? and why is 9 so high? <<
Agreed, the data doesn't quite fit my theory. I keep waiting to see #9 jump to #4
| 1:15 am on May 22, 2008 (gmt 0)|
In my niche...
I am suddenly no 4.... for some time of the day.... and other times not...
What I see here is the definitive number 1 has the niche KW as his domain name... So G figures... that's solid. It belongs in number 1 beause it's spam-proof. The domain IS the KW... His products are the KW... hence number 1
Number 2 is the same guy as number 1 except it's a secondary page and again, the KW is part of the page URL... another safe bet.
Number 3 is WIKI
And... I'm number 4 along with others in the same niche.
Can this possibly be not G testing the number 4 position but just simply rotating the algo so visitors see different niche players?
Also... a rotating algo is harder to spam and easy to employ.
| 5:13 am on May 22, 2008 (gmt 0)|
Thanks for all the analysis. When you look at the pages indexed and the backlinks it doesn't even make sense. No rhyme or reason to anything.
In my topic some of the things that are popping in and out of positions 4 to 6 are so odd, I can't imagine how they could be ranking like they do.
| 5:49 am on May 22, 2008 (gmt 0)|
|So what do you think? Is size of site a potentially big ranking factor? |
I don't think you can generalize. The size of the site often correlates with some factors that will boost it up in rankings, but if badly done, size can backfire. In some niches, largeness can suggest importance, and that can attract high quality inbounds. More good content can attract more inbounds from more sources.
But many sites expand beyond what their inbounds can support, or they may not be structured well. In these cases, size can work against them. I remember a recent discussion somewhere on WebmasterWorld about how reducing the number of pages in a site increased rankings. It was something that didn't surprise me.
Re backlinks from Yahoo, numbers don't always tell the story. Very often, large numbers are due to low quality ROS inbounds... and a relatively few high quality inbounds could knock those out of the water.
Large size might also suggest a large company, and that might be an advantage for certain phrases... perhaps a disadvantage on others.
| 4:37 pm on May 22, 2008 (gmt 0)|
does this happen for any type of search someone does? i haven't done as much research into this as others but...
are the top three in the serps any type of website or has anyone seen them to typically be a wikipedia page, amazon, ebay, blog, linkedin, news site, .edu, .gov or something that's not solely a url like joeswebsite.tld?
could the top three spots be reserved for a special type of site and normal websites (joeswebsite.tld) be now reserved for spot 4 and below?
(i hope i made sense)
| 5:46 pm on May 22, 2008 (gmt 0)|
The top 3 for the single search term I've been watching has stayed the same almost the full time the top serps have been jumping around. But I think it depends on the search.
My theory is that those items at the top are just so much more solid in terms of size and links that they aren't affected by this.
It is annoying that the poorly written Wikipedia page for my search term is number three when there are so many excellent sites on the topic that should be there. But that's the way is seems to go, Wikipedia rules.
| 6:31 pm on May 22, 2008 (gmt 0)|
Is that because it is Wikipedia or because of its structure. Internal linking or external linking?
I remember during the Florida aftermath we noticed directories and search scraper sites rise up the serps it turned out that this was, in part, because outbound links with the right anchor text were given extra weight. So it wasn't because they were directories but because of what directories happen to do, ie outlink with on topic anchor text.
Could there be something like this hidden in the statement "Wikipedia rules"?
| 7:48 pm on May 22, 2008 (gmt 0)|
I think it is in Wikipedia's linking structure. They are just doing things right. But you have to be a big system for it to work. Sites like About have a similar advantage.
| This 77 message thread spans 3 pages: < < 77 ( 1 2  ) |