I have been testing some stuff, basically when we got a visitor to a particular site, I checked their IP, opened a proxy and found my search position for the terms they had landed on the site for, results follow

Rankings  1-10       3975 hits    70.6 %
Rankings 11-20      572 hits      10.2 %
Rankings 21-30      229 hits      4.1 %
Rankings 31-40      244 hits      4.3 %
Rankings 41-50      100 hits      1.8 %
Rankings 51-60      30 hits        0.5 %
Rankings 61-70      17 hits        0.3 %
Rankings 71-80      17 hits        0.3 %
Rankings 81-90      45 hits        0.8 %
Rankings 91-100    42 hits         0.7 %

the last time I did this people were not digging so deep into serps, poor quality results I Guess ?

DaveN

DaveN

10 Comments

  • 1

    I’ve seen some go as far as the 400-500 range to hit some of my (quality) sites in particularly spammy neighbourhoods. It’s interesting and probably spot on about the quality of the SERPS. I used to find the ones that made the most effort were looking for specific porn niches, and obviously didn’t mind looking through 300+ sites to get there – and would have been disappointed hitting my sites.

    Dio | http://thenextbigwebthing.com

    21st April 2007 @ 15:02

  • 2

    Dave

    Any chance that’s bot traffic or zombies?

    Even if I accepted the 11-20 results as possible, the rest of the numbers don’t follow a predictable curve.

    Meanwhile, it would be fantastic to chart that data against the number of terms in the query.

    I have heard recently that one of the new “markets” for zombie armies to have them dig deep into results and drive clicks to particular sites under the assumption that user data is helping determine rankings.

    Jonah Stein | http://www.alchemistmedia.com

    21st April 2007 @ 18:29

  • 3

    Dave -

    You may be interested in comparing your data with an analysis of the 20 million clickthroughs from the AOL data http://www.seo-blog.com/serps-position-and-clickthroughs.php

    Also compare your 1-20 with http://www.seo-blog.com/position-and-clickthrough-tool.php

    - Michael

    Michael | http://www.seo-blog.com/

    21st April 2007 @ 22:01

  • 4

    I don’t see this at all going on with my SEO blog but its an everyday kinda thing with my porn-related sites. I get hits from 700-800 range sometimes. Usually it means the first few hundred results are all crap.

    I see people switching from regular SERP to Google Images when looking for porn. Interestingly, alot of adult webmasters don’t know how to optimize for Google Images or are not allowed to do so because of rules dictated from above (LL/TGP owners).

    Halfdeck | http://www.seo4fun.com/blog/

    21st April 2007 @ 23:58

  • 5

    On a site with just 17% organic traffic. Entertainment / humour niche
    1-10 656 75.80%
    11-20 108 12.50%
    21-30 39 4.50%
    31-40 21 2.40%
    41-50 6 0.70%
    51-60 6 0.70%
    61-70 3 0.30%
    71-80 4 0.50%
    81-90 1 0.10%
    91-100 1 0.10%

    On a site with 82% organic traffic – more technical niche
    1-10 10291 90.80%
    11-20 538 4.70%
    21-30 174 1.50%
    31-40 74 0.70%
    41-50 34 0.30%
    51-60 23 0.20%
    61-70 18 0.20%
    71-80 7 0.10%
    81-90 5 0.00%
    91-100 3 0.00%

    Liam Victor | http://www.liamdelahunty.com/tips

    22nd April 2007 @ 01:14

  • 6

    Jonah – zero chance that’s bot traffic. Robots never send Referer headers.

    Rob Haswell

    23rd April 2007 @ 09:40

  • 7

    Rob – yes, but they could. easily.

    Stefano | http://www.diarizing.com

    24th April 2007 @ 02:00

  • 8

    It could possibly be some bots–but as we see more personalized search results, there could be differences. I’m seeing more and more people use custom setting such as listing 100 results at a time rather than the default 10.

    Bill | http://www.billhartzer.com

    26th April 2007 @ 23:10

  • 9

    @Rob Some bots send a referer (Live sens a bad referer), a few even send a correct referer. Mine do. There is as many bots sending a correct referer than there is browsers not sending a referer (because of a proxy, norton, …).

    da108ed4

    28th April 2007 @ 12:35

  • 10

    It pretty much looks like an exponential drop, from page 2 onwards. I don’t think it will ever change the distrubution will be exponential drop – not including the robots.

    Haniff Din | http://www.haniff.co.uk

    30th April 2007 @ 19:34

Write a Comment

*

*

*

SES New YorkA4U Expo Munich
Subscribe
to the David Naylor feed
Follow
David Naylor's Twitter feed

View Dave's Blog