|A SERP's experiment|
| 7:06 pm on Oct 1, 2012 (gmt 0)|
For months now I have been tracking one specific set of keywords in regard to an e-commerce site.
First, and in comparison, using DuckDuckGo, the page ranks #3 in the results after 'A' and 'e'. In Bing it's 5th with 'A' being 1 & 3, and 'e' being 2 & 4 which makes this the first non-big/sell all site in the results. This has been consistent for about 4 months with only a very minor fluctuation, up/down one or two places. Then there is Google. Below is a four month study of the same result from them.
9/30/2012, 9:20am, Page 19, #3
9/27/2012, 3:10pm, Page 22 #8
9/22/2012, 4:55pm, Page 3, #6
9/17/2012, 12:25pm Page 3 #7
9/14/2012, 10:25am Page 15, #5
8/21/2012, 9:00pm, Page 2 #10
8/1/2012, 10:30am, Page 4 #10
7/20/2012, 1:00pm, page 16 #10
6/28/2012, 11:25am, page 11 #4
6/25/2012, 11:45am, page 10 #7
6/13/2012, 11:30pm, page 9 #8
6/12/2012, 4:30pm, page 9 #10
6/9/2012, 7:30pm, page 11, #2
6/9/2012, 9:30am, Page 6 #5
6/8/2012, page 11 #4
6/7/2012, page 10 #4
6/6/12, 10:30am, page 11 #3
6/5/12, 1:30am, page 5, #1
5/18/12, 11:20, page 4 #6
5/16/12, 11:00, page 7 #4
5/14/12, 2:45pm, page 7 #7
5/10/12, 11:55pm, page 3 #10
5/9/12, 9:19pm, page 3 #8
It should be noted that the only changes to the page was the quantity of an item or deletion of an out of stock item.
And on a side note, but not unrelated, their claim that they are going to reduce the number of consecutive results from the same site is laughable. 'A' has over 30 results in the first four SERP's with 24 listings in a row. How is this fair to other retailers.
All things considered, does this make any sense? Frankly, when it comes to Google, I just feel like giving up.
| 1:23 pm on Oct 2, 2012 (gmt 0)|
Only to a highly volatile machine learning algo - with lots of testing, geo-location and personalization.
Day-to-day ranking reports for Google are pretty much an antique way of doing SEO for Google today, IMO. I've been seeing this evolution for a while, and now I see it in spades. The primary SEO metrics need to be about traffic and conversion. Ranking positions are a diagnostic that we might use as part of understanding sudden traffic changes.
We can't go into the future expecting Google to be what it was in the past. It's not going to happen.
| 3:55 pm on Oct 2, 2012 (gmt 0)|
The ranking reports may be an antique way of measuring anything, but I found the results more than curious. For what it is worth, the page url is manufacturer/product_line which are the three words I used for the test. Even more curious, maybe, is it use to rank in Google as it does in Bing and DuckDuckGo. The page does not use any black hat SEO and does not contain irrelevant information. In fact, it doesn't even use the manufacturer's standard description, but instead a combination of that and original content. Bottom line, I do not know what to think other than you can crazy trying to figure it out.