While the world is looking at users in AOL search data I aggregated it a little.
There has been some discussion on how much #1 place in SERPs really matters. Some say it doesn’t matter so much as long as you are on the first page while others claim most of the search traffic goes to first places. Since we have the AOL search data we can do a little research.
Where do people click in AOL SERPs?
Number of user clicks per page
| Results page | User clicks |
| 1. page | 17429435 |
| 2. page | 854703 |
| 3. page | 416025 |
| 4. page | 208950 |
| 5. page | 111607 |
| others | 413820 |
90%! Wow! This is a lot more than I have expected. If one wants to see some search engine traffic his website must be on first page for desired search keyword.
Number of user clicks per position
| Results position | User clicks |
| 1. place | 8220278 |
| 2. place | 2316738 |
| 3. place | 1640751 |
| 4. place | 1171642 |
| 5. place | 943667 |
| 6. place | 774718 |
| 7. place | 655914 |
| 8. place | 579206 |
| 9. place | 549196 |
| 10. place | 577325 |
| others | 2005105 |
First is best by far. Almost half of the users clicked on the first result.
Based on the revealed AOL search data we can conclude that highest positions in search results receives almost all the traffic. Winner takes it all.
It would be nice to see similar statistical breakdown with thematic of the search query (brand, health, technology, …) included. It is well known that in some fields people dig deeper and go further…
11 comments
Comments feed for this article
Trackback link
http://davors.eu/2006/08/aol-90-clicks-on-first-page-4-on-second/trackback/
August 17, 2006 at 4:23 am
Pingback from Sanjeev.NET » Blog Archive » links for 2006-08-17
August 21, 2006 at 8:28 pm
Pingback from AOL statistika, reddit.com in dva tisočaka. « rufbetten [moje postelje]
August 17, 2006 at 1:18 am
Jawad Shuaib
Very good analysis. This also hints us a bit towards ads placed on search engines. Most search engines (i.e, Google, Yahoo, Ask, AOL) don’t reveal how many clicks third party ads are receiving, however, with analysis like these, we can get a rough sketch of it.
Jawad Shuaib
August 17, 2006 at 2:59 am
Stephane Grenier
Hi,
Your data is very interesting and very nicely presented. I especially like the graphs!
I recently read an article at SeoBlackHat.com (http://seoblackhat.com/2006/08/11/tool-clicks-by-rank-in-google-yahoo-msn/) where he ended up with very similar statistics from the AOL data. You might be interested to check it out as he also compares the AOL data with other search engines using the Overture tool, etc.
Regards,
Stephane Grenier
Founder
LandlordMax Property Management Software
.http://www.LandlordMax.com
.http://www.FollowSteph.com
August 17, 2006 at 12:15 pm
Geir
Interesting results. I wonder if one should differ between searches where the user is searching for a specific page or just searching for information on topic. I often just type “blabla wikipedia” in google instead of first opening wikipedia and then use the search feature there. In most cases the page I’m looking is on top.
August 17, 2006 at 1:16 pm
ruph
Geir Yep, I agree. I thought about it to, but it is a lot harder to achieve that..
It would be a lot more informative if I could somehow differ between users searching for a domain (a lot of people is searching domain name in search engines instead just entering it in adress bar) and others…
August 17, 2006 at 5:30 pm
Pat Rock
One thing that is coming out (and you can easily check this) is that its better to be #1 on page 2, than #10 on page 1. (in aol dataset search for rank 11).
Pat
August 17, 2006 at 8:22 pm
ruph
Pat Rock This is not correct..
… 1. page ….
8. place = 579206
9. place = 549196
10. place = 577325
…. 2. page …
11. place = 127688
12. place = 108555
Second page is clearly second. But it is better to be #10 than #9.
#9 is the unlucky one…
August 17, 2006 at 8:54 pm
Matt
Great analysis, thanks. I would point out that I think it would be highly dangerous to apply these types of numbers to search behavior on Google or Yahoo. AOL customers are highly unique web users and behave very differently than standard ISP users.
But great info nonetheless.
August 23, 2006 at 9:39 am
MaKa
@ Pat
Why wouldn’t it be correct that #10 is more popular than #9. It makes sense to me. People scroll down to got to the next page selector and they either spot #10 and change their mind or simply click on #10 by accident while scrolling down.
April 9, 2011 at 3:39 pm
søkemotoroptimalisering
I personally believe that it is better to be on 10. place than 9 because it is more visible position, and from my experience, this position receives more clicks.