Are you a Google Analytics enthusiast?
Share and download Custom Google Analytics Reports, dashboards and advanced segments--for FREE!

www.CustomReportSharing.com
From the folks who brought you High Rankings!
More SEO Content
International SEM | Social Media | Search Friendly Design | SEO | Paid Search / PPC | Seminars | Forum Threads | Q&A | Copywriting | Keyword Research | Web Analytics / Conversions | Blogging | Dynamic Sites | Linking | SEO Services | Site Architecture | Search Engine Spam | Wrap-ups | Business Issues | HRA Questions | Online Courses
A/b Split Testing
Started by
JustLearnin
, Apr 16 2004 12:07 AM
6 replies to this topic
#1
Posted 16 April 2004 - 12:07 AM
I am curious about the best way to run a split test of two headlines (etc)
online, with all the rest of the body copy the same ..... search engines
dont like duplicated content, so is there a legit way to accomplish this and
get the info without having it become a SE spamming issue?
Any thoughts would be appreciated .....
online, with all the rest of the body copy the same ..... search engines
dont like duplicated content, so is there a legit way to accomplish this and
get the info without having it become a SE spamming issue?
Any thoughts would be appreciated .....
#2
Posted 16 April 2004 - 12:14 AM
Run two pay per click ads on google (using the same words and ad copy) with each directed to a different web page (with different headings). make sure neither page is linked to from any other page on the web...
there are ways to dynamically code the pages so that they show different stuff for different visitors too, but the first is way easier in my opinion...but i do not program by trade...at least not yet
there are ways to dynamically code the pages so that they show different stuff for different visitors too, but the first is way easier in my opinion...but i do not program by trade...at least not yet
#3
Posted 16 April 2004 - 07:55 AM
Yep, use PPC ads to test it.
But not only should you make sure the landing pages aren't linked from other pages of your site, you should exclude them from being spidered through the robots.txt exclusion file.
Traditional SEO is not very good as a testing ground for copy.
Jill
But not only should you make sure the landing pages aren't linked from other pages of your site, you should exclude them from being spidered through the robots.txt exclusion file.
Traditional SEO is not very good as a testing ground for copy.
Jill
#4
Posted 16 April 2004 - 07:03 PM
Thank you both ... I can see how the PPC would be safest from an SEO
perspective ..... and the robots.txt exclusion makes sense, too ....
If I wanted to dynamically serve up the pages like this on my site, instead
of using a PPC:
somesite.com/index.html
where index.html has the ability to serve up test-1.html amd test-2.html for the split test without people being aware of index.html ....
There is no way to not have SOME kind of incoming link (unless preexsiting
keyword search engine traffic comes in) to get people there ....
If I have already been spidered by the SE, could I run this setup (if I already had
decent traffic) for a few weks to see what the result was and stay out of trouble?
Or .... maybe not use the front door page, but instead:
somesite.com/test/index.html with test-1.html amd test-2.htm in this "test" directory ....
and a link on the website main page that leads into the test directory ..... would this kind of link
be the kind of thing you are talking about avoiding?
And, if it is ad copy leading to a direct sale, there would be an outgoing link from both test pages ..... so .....
I guess I am trying to figure out if there is any way to do it dynamically onsite without PPCs
without causing SE problems ... either by the way it is setup, or the length of time
it is allowed to exist ....
Or not :-)
perspective ..... and the robots.txt exclusion makes sense, too ....
If I wanted to dynamically serve up the pages like this on my site, instead
of using a PPC:
somesite.com/index.html
where index.html has the ability to serve up test-1.html amd test-2.html for the split test without people being aware of index.html ....
There is no way to not have SOME kind of incoming link (unless preexsiting
keyword search engine traffic comes in) to get people there ....
If I have already been spidered by the SE, could I run this setup (if I already had
decent traffic) for a few weks to see what the result was and stay out of trouble?
Or .... maybe not use the front door page, but instead:
somesite.com/test/index.html with test-1.html amd test-2.htm in this "test" directory ....
and a link on the website main page that leads into the test directory ..... would this kind of link
be the kind of thing you are talking about avoiding?
And, if it is ad copy leading to a direct sale, there would be an outgoing link from both test pages ..... so .....
I guess I am trying to figure out if there is any way to do it dynamically onsite without PPCs
without causing SE problems ... either by the way it is setup, or the length of time
it is allowed to exist ....
Or not :-)
#5
Posted 16 April 2004 - 10:35 PM
None of that will get you in trouble with the search engines.
Jill
Jill
#6
Posted 22 April 2004 - 12:42 PM
excellent .... always good to hear the real scoop instead of me floundering around and wondering!
Thanks ....
Thanks ....
#7
Posted 26 April 2004 - 08:15 PM
Hello Everyone!
In his newsletter, Ebook Times, Michael Hopkins had an article about using a web-based app called Scientific Internet Marketing Assistant (SIMA) to do exactly this type of testing. I've never tried this app but I was reading about it earlier tonight and then saw this thread.
You can see the article here [http://bizzydays.com...b_2004.htm#tip]
John
[Edited live link per forum guidelines -Randy]
In his newsletter, Ebook Times, Michael Hopkins had an article about using a web-based app called Scientific Internet Marketing Assistant (SIMA) to do exactly this type of testing. I've never tried this app but I was reading about it earlier tonight and then saw this thread.
You can see the article here [http://bizzydays.com...b_2004.htm#tip]
John
[Edited live link per forum guidelines -Randy]
Edited by Randy, 27 April 2004 - 07:00 AM.
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users








