Welcome to WebmasterWorld Guest from 23.22.182.29

Forum Moderators: phranque

Message Too Old, No Replies

click distribution

how user clicks are distributed depending on the position on a search resul

     
7:13 pm on Jun 17, 2003 (gmt 0)

New User

10+ Year Member

joined:June 17, 2003
posts:11
votes: 0


Hello,

Does anyone know how user clicks are distributed depending on the position in search results? What I mean is a table that would list the percentage of the users that are likely to click the first, second, etc result of a search.

Or in other words, if we have the 10th position and getting 1000 visitors per day for a particular keyword, how many visitors we are likely to get if we are placed 1st for the same keyword?

Thanks

8:16 pm on June 17, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Mar 10, 2000
posts:2151
votes: 0


I've never seen any definitive information but based on my experience #1 and #2 are killer (obviously) then it settles down through the middle and #9 and #10 do better. Kind of an upside down bell-curve. I recently had a #6 slot for a very competitive term and it hardly turned any sales compared to when I was sitting #9 or #10.

btw - welcome to WebmasterWorld nicksimon

8:29 pm on June 17, 2003 (gmt 0)

New User

10+ Year Member

joined:June 17, 2003
posts:11
votes: 0


Thanks, Oilman

What I am after is a return on investment analysis. We spent some significant effort moving our site from #7 to #4 and it looks like it resulted in a 50% increase in the related traffic.

Now the question is what kind of improvement are we to expect by moving from #4 to #2 or #1. Is it 50, 100, 300%?

I know what to do and what kind of investment is needed to get there, but am not sure about the rewards.

8:39 pm on June 17, 2003 (gmt 0)

Senior Member from US 

WebmasterWorld Senior Member ken_b is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month

joined:Oct 5, 2001
posts:5667
votes: 60


Nicksimon;

it resulted in a 50% increase in the related traffic.

If I may ask (feel free not to answer this), but was there a corresponding increase in conversions?

Or did it just result in more traffic?

10:39 pm on June 17, 2003 (gmt 0)

New User

10+ Year Member

joined:June 17, 2003
posts:11
votes: 0


>>If I may ask (feel free not to answer this), but was there a corresponding increase in conversions?

I don't know yet. We've been #4 for 2 weeks, but as we sell shareware and most people only pay when the program stops working after the evaluation period is up - 45 days in our case - we have to wait for another month or so for any increase of revenue.

Also, the 50% I was talking about was only from GG on the exact keyword search. The overall traffic is practically the same, however the number of downloads is up about 30%. That shows how effective the SE are. In our case SE account for about 60% of traffic and 95% of revenue

10:42 pm on June 17, 2003 (gmt 0)

Preferred Member

10+ Year Member

joined:June 11, 2003
posts:417
votes: 0


nicksimon

I don't know of any resources/tables - but I'll share what I know anyway!

Getting from position 9/10 up to positions 1-3 would be worth a lot of effort. This stops a page 2 competitor creeping up on you and (during the next update) overtaking you. The difference between any first page listing and appearing at 11 on page two is huge - we really noticed the difference when this occurred.

Spending effort/investment moving from position 4 to 2/1 is not (in my opinion) worth the investment (unless the investment is quite small). There are too many "what ifs" for this to be worth the effort unless you've got so much money you don't know what to do with it!

I say this because I wouldn't know if:
1. The SERPS will be ranked in the same way (new algorithms are frequent enough)
2. If competitors aren't trying to do the same (i.e. the ones already above me in particular). It may be extremely difficult to overtake them anyway (look at the number/quality of their backlinks - sometimes it can be an order of magnitude to get above them)
3. It may not be necessary - perhaps the title/description is good enough that people will click on your site instead of one of the sites listed above.
4. There may be better ways to invest time and money - e.g. target other popular keyterms and target them, start doing PPC/affiliate programs/email marketing/target other engines/improve conversion ratio (usability etc). These could see higher return on investment while making the site more robust and much less dependent on a few search terms.

10:58 pm on June 17, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:May 14, 2002
posts:1192
votes: 0


TravelSite,

Welcome to WebmasterWorld!

> 3. It may not be necessary - perhaps the title/description is good enough that people will click on your site instead of one of the sites listed above.

That is an excellent observation! Too many people try to stuff their title with keywords, forgetting that Google will never buy anything from them :)

An attractive title in the top half of the page can easily beat the #1 site if the latter has an unattractive title.

To attempt to answer nicksimon's unanwerable question :) I will suggest that the top half of the first page is better than the bottom half. The chances are very high that most searchers will find something attractive to click on in the top half, and if it answers their question they may look no further.

10:42 am on June 18, 2003 (gmt 0)

Moderator from AU 

WebmasterWorld Administrator anallawalla is a WebmasterWorld Top Contributor of All Time 10+ Year Member

joined:Mar 3, 2003
posts:3701
votes: 3


In many SERPs I see, including my own terms, the page is full of irrelevant results, so I don't think it is possible to say absolutely that #1 is better than #2. I have seen others say that they do better at #2 than #1 and I think they mean that both are truly competing hits.
10:55 am on June 18, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Mar 25, 2003
posts:970
votes: 0


Also who you are competeing against plays a factor. Are the people above you giving Good product or services like yo do. If yes then ranking 3 may also not help you. If the people you are competing against have pathetic services or their website looks disgusting or are irrevelant results than your current ranking is not so bad. A better way is to study the psychology of the potential customers(searcher) and try to bring answers to their queries in the visible part of SERPS.

One interesting observation i read was from digitalghost who said something to the effect that ranking on no 11 gives more traffic then on no 9 and 10.

Now the question is what kind of improvement are we to expect by moving from #4 to #2 or #1. Is it 50, 100, 300%?

Unfortunately this is a not a simple matter of statistics. HTH :)

1:21 pm on June 18, 2003 (gmt 0)

New User

10+ Year Member

joined:June 17, 2003
posts:11
votes: 0


Thanks everybody for your input. However there is no answer to my question.

Of course a good description is better than a bad one, or a relevant result is more likely to get clicked. That was not my question.

Let's be scientific and compare apples to apples. To analyze a factor you have to assume or other factors being equal.

In or case, the question really is:

How the users clicks are distributed when all the results are equally relevant and have a similar title/description?

Or, similarly:

If in the list of results my listing is #4 and, say 1000 users a day click on it, how many users will click on the listing if it is moved to #1 position and the old #1 becomes #2, #2 - #3, and #3 - #4 with nothing else changed?

Have anyone measured, or at least can estimate, how their traffic behaved when the position in SE shifted between #1 - 10 with all other relevant components being the same?

1:33 pm on June 18, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Feb 25, 2003
posts:972
votes: 0


nicksimon, I know exactly what you're looking for and I've got an Excel sheet with 162 data points trying to do it.

About 20% of the way into it I had a 1/x type relationship that looked pretty good and I thought I had it figured out. However, after adding another 120 or so data points, it just looks like somebody took a Sharpie and started spattering dots all over my chart. I've tried filtering the data numerous ways but there's not a formula to dig from it.

I was basing it on the percentage of clicks from Google using a mix of Word Tracker and Overture search term popularity numbers. I also know for a fact that I'm not the only one who has tried to find that 'formula' you're looking for but everyone I know who's tried to find it has decided that the above posts are correct....

The only 'formula' for position vs CTR is higher on page one gets more clicks.

BTW, it would be almost impossible to get rid of the other factors such as title to get a pure set of data. You would need to have control of both Google and the other sites in the top ten.

2:02 pm on June 18, 2003 (gmt 0)

Preferred Member

10+ Year Member

joined:June 11, 2003
posts:417
votes: 0


You could always set up a controlled experiment yourself. E.g. set up a mock html page with similar titles etc - then get lots of friends to sit down and see what they click on first (don't tell them thats what your studying).

You could then take this a stage further by changing various aspects of the page (ppc advertising included, changing titles etc) to see what effects it has.

If you're considering it for PPC then you may get help. If your client/company is wanting to see what difference in traffic there is between the top 10 spots (to see if its worth spending more money on) then a PPC company may have figures for you (it couldn't do any harm to ask).

Perhaps someone in the Pay Per Click Engines forum could help with stats similar to what you need.