Welcome to WebmasterWorld Guest from

Forum Moderators: open

Message Too Old, No Replies

The raw statistics - Ranking in Google

SERP in relation to PR and Click through rate.

7:16 pm on Nov 3, 2002 (gmt 0)

Junior Member

10+ Year Member

joined:May 30, 2002
votes: 0

Sorry if this subject has appeared at one point or another previously on WebmasterWorld, but has anyone got information on click throughs and PR in relation to SERP from Google? It would really be interesting to see the exact values (of which maybe only really Google knows).

This is mostly guess work and probably inaccurate, but I've tried myself to estimate (and research a little for the PR value) the above. Also, I've listed two 'tables' - one for popular keywords, and another for niches. I suppose one could split the tables into even more categories (such as surfers with the intention of shopping etc.), but I'll stick at this for now.

Alongside all these results, the average PageRank estimation of that site positioned at xyz SERP would be very interesting too. This would then indicate how much PR is factored into the equation generally. Once again, this is a little guesswork (mixed with a bit of research this time) on my half - so the real results could be fairly different.
(Bear in mind that all percentages for click throughs will exceed 100%, since the visitor is often inclined to visit more than one site. In other words, it represents the chance of a click through)

Mega-popular 1 or 2 keyword phrases:

SERP 1: 40% chance of that visitor visiting your site if ranked 1st in Google. PR 8.8
SERP 2: 30% chance if ranked 2nd. PR 8.6
SERP 3: 24% ... PR 8.5
SERP 4: 20% ... PR 8.4
SERP 5: 16% ... PR 8.4
SERP 6: 13% ... PR 8.3
SERP 7: 11% ... PR 8.3
SERP 8: 10% ... PR 8.3
SERP 9: 9% ... PR 8.2
SERP 10: 8% ... PR 8.2
SERP 11: 4% ... PR 8.1
SERP 15: 3.8% ... PR 7.7
SERP 19: 3.5% ... PR 7.4
SERP 20: 3.0% ... PR 7.4
SERP 21: 2.4% ... PR 7.4
SERP 25: 2.1% ... PR 7
SERP 29: 1.9% ... PR 6.8
SERP 30: 1.3% ... PR 6.8
SERP 40: 0.7% ... PR 6.6
SERP 50: 0.4% ... PR 6.5
SERP 60: 0.25%... PR 6.4

....... with an approximate halving of click throughs for every 10 places further down perhaps?
Maybe I've got it wrong - perhaps the real results form a different kind of 'curve'. Could there be a formula for this?

Incidentally, approximately 1-2 out of every 100 sites were given (whether deserved or undeserved) the dreaded PR 0. This might even be because the pages are new, so it's wise not to jump to any conclusions. Also, this figure could vary widely for other searches.

Anyway, vastly different results for obscure, and niche keywords - are as follows:
SERP 1: 30% chance. PR 6.6
SERP 2: 27% ... PR 6.45
SERP 3: 24% ... PR 6.3
SERP 4: 22% ... PR 6.1
SERP 5: 19% ... PR 6.0
SERP 6: 17% ... PR 5.9
SERP 7: 15% ... PR 5.8
SERP 8: 13% ... PR 5.7
SERP 9: 11.5%.. PR 5.7
SERP 10: 10% . PR 5.6
SERP 11: 8.5% . PR 5.5
SERP 15: 7.5% . PR 5.3
SERP 19: 7.2% . PR 5.1
SERP 20: 5.7% . PR 5.1
SERP 21: 5.4% . PR 4.9
SERP 25: 5.2% . PR 4.7
SERP 29: 5.0% . PR 4.5
SERP 30: 4.3% . PR 4.5
SERP 40: 3.4% . PR 4.4
SERP 50: 2.6% . PR 4.2
SERP 60: 1.6% . PR 4

....... with an approximate halving of click throughs for every 15 places further down perhaps? Once again, has anyone obtained more accurate information (for even further rankings below SERP 100 too) or a formula for this?

7:23 pm on Nov 3, 2002 (gmt 0)

Full Member

10+ Year Member

joined:June 11, 2002
votes: 0

Hi dwithe, nice work if it is accord to the reality. wich method have you used to get this tables?
1:09 am on Nov 4, 2002 (gmt 0)

Junior Member

10+ Year Member

joined:May 30, 2002
votes: 0

Well the PR score was fairly easy to calculate, but took a long time to do (wade through SERPs and establish an approximate average).
The click through rate (percentage) though is more an 'educated guess' than anything else - which is why I was hoping others might have more accurate data.
I believe even the much sought after PR 'log base' could be derived with even more research from SERPs in relation to PR value.
1:36 am on Nov 4, 2002 (gmt 0)

New User

10+ Year Member

joined:June 25, 2002
votes: 0

I would expect position 11 (and 21..) to generate more clicks than position 10 (and 20..).

It may sound counter-intuitive, but there's something inherently more appealing about 'above the fold' information.

Pick up a newspaper and notice how the top of page 3 is more eyecatching than the bottom of page 2 - we've just been trained that way.

3:35 am on Nov 4, 2002 (gmt 0)

Senior Member

WebmasterWorld Senior Member chiyo is a WebmasterWorld Top Contributor of All Time 10+ Year Member

joined:June 21, 2000
votes: 0


It takes into account and highlights that behaviour of searchers who do niche searches is very different from popular terms. It seems you are saying they are more likely to look down the list to click, rather than to click the top one or two. I would guess they also go back to the SERP to click more.

4:13 am on Nov 4, 2002 (gmt 0)

Preferred Member

10+ Year Member

joined:May 1, 2002
votes: 0

Not for nothin' but I'd like to see the support for this. Many of your premises seem faulty. For one thing how do you judge the popular vs. the niche? Maybe most importantly, equating PR with SERPs is not often the way it plays out. I know from my own experience with a PR 6 - 7 that many sites with 4s and 5s rank higher on certain combos. I just checked a very popular search combo and found a bunch of PR6s above a PR8.

Also, as a final note, I'm in a heavily trafficed topic - genealogy - and I do quite well on page two of serps. I don't claim to be any sort of expert. I get no more than 2,000 referrals from Google, proper per day but I don't see the trends you are trying to demonstrate.

So I have to ask what is the source of your data?

7:10 am on Nov 4, 2002 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:May 21, 2001
votes: 0

I don't have any of my stats from where I am now, but two comments:

1) The percentages are too high

2) The PR differences are probably SOMEWHAT RELATIVELY accurate. PR ranges dramitically from term to term. It is almost useless to come up with an average - as you do not need a high PR to get placed for "fuzzy wuzzy blue green widgets", nor can most people get the PR needed to place highly for music.

People need to do more research on THEIR own keywords in their industry. Their are many keywords out there that are underused and profitable. The basic strategy that companies like wordtracker suggests DOES WORK.

Just my 2 cents

1:00 am on Nov 6, 2002 (gmt 0)

Full Member

10+ Year Member

joined:June 11, 2002
votes: 0

Chiyo, I'm not agree with your guessing. A computer screen is not a newspaper and a guessed "rational user" should finish all the options before to click to second page. In addition, I would set three categories:
1. Most popular KW (#1,#2 and #3 are the winners)
2. 3 KW combos (you need more clicks)
3. More than 3 words (not so many clicks)

This is infered studiying my own searchs so it's not conclusive but when I search for a "walt disney interview" good results are much more hard to find than "walt disney interview new york 79".
My 2 cents