- Test ..
- When writing a sales letter (or design anything for that matter) you can (and will) have many options or alternatives without exactly knowing how things work out in practice.
Therefore you have to TEST.
I give you an example. Which headline do you think is best:
-How to make chocolate pudding in 6 minutes.
-Tonight serve this ready-mixed chocolate pudding.
Well?
Ok, the second one is better. How I know that? It was tested!
This example comes from a really great book called "How to Make Your Advertising Make Money" by John Caples, vice president of BBD&O and master of Direct Marketing.
ISBN: 0-13-423608-4.
I think it's available on the Internet at Books.com, now owned by Netmarket.com.
Another title by the same writer is "Tested Advertising Methods", ISBN0-13-244609-X.
Great stuff!! If I remember correctly I paid about 10 dollars for each. That's a steal.
Highly recommended even if you decide not to go online!
(These books and many others are recommended by Marlon Sanders in The Amazing Formula).
OK, let's get on with our subject.
- And Test ...
- The good news.
The good news is that Internet is definitely by far the cheapest and fastest way to test sales letters, products, classifieds, banners, catalogs, brochures etc.
And that makes life easy.
You can run a sales letter one or two weeks and measure the results in visitors AND sales. Than you replace the headline and test again during the same period of time. You can do this with every detail of your web site. Leave the parts that deliver the best results and replace the parts that give bad results. In no time you shall have the best combination possible.
- And Test Again!
- The message is: TEST, TEST, TEST!
But how can we find out the results? That's done by what is called Tracking.
Every visitor leaves some trails on your website that are logged in a logfile and can be viewed.
NO, NO, I cannot know who you are, where you live, if you are married, number of kids or income. IMPOSSIBLE!
Log files can deliver information like:
- the visitors IP number, that is the number their ISP assigned to the visitor when he/she was connected to the Internet;
- where the visitor came from, the site he/she visited before visiting yours;
- the keyword(s) the visitor used to find you;
- the visited pages and time spend;
- where the visitor goes to when he/she clicked on a link at your site.
Special statistics programs add log file data so it can show:
- total time spend by visitors per page
- total number of visitors per day/week/month
- grand total of visitors.
- Statistical Info
- This makes it possible to see the visiting results of certain pages on your site over a period of time.
And if you count the sales results over the same period, you have an excellent way of measuring the effectiveness of what ever you are testing.
CGI-scripts or special programs can perform log file tracking.
Professional scripts can be expensive.
But I promised you a test site for almost zero dollars. And you have already spend 17 US dollars buying MYSS, so to keep my promise I will give you some excellent FREE resources for tracking scripts.
Here they are:
AXS by Fluid Dynamics at:
http://www.xav.com/scripts/axs/
This is a server-side CGI-script that you have to install!
Analog by Stephen Turner at:
http://www.statslab.cam.ac.uk/~sret1/analog/
You can install this program either server side or client-side. No need to know about programming! Just Windows and downloading your logfiles with FTP.
But you still need to install these programs. That might be a little difficult (although Analog is quite easy).
So here's the ultimate alternative.
A site that does the tracking for you. You only have to copy a piece of HTML to your website.
Its name is Extreme Tracking and you can find it at
http://www.extreme-dm.com/tracking/
Be aware that everyone can see the results, but all in all it's a great service for a test site.
An alternative for that is
http://www.hitbox.com/
- Sales versus visitors
- OK, now you know the big picture about testing, tracking and internet statistics.
It is said that the total Internet average is that about 1 visitor out of every 150- 200 will buy.
And this is likely going to be 1 out of every 300.
You've seen the home page of MYSS.
It's an excellent example of writing sales letters and it's tested over and over again. Because I'm a member of the "MYSS-family" I know that about 4 or 5 out of every 100 visitors buy. And at special occasions (special promotions) that figure can go up to 9 out of every 100!
Now that is 13 to 18 times more than average! That is what tracking and stats can do. Don't you want that result too?
See how MYSS works!
If you can't solve problems just send an Email to:
Warm regards,
Case Stevens
Moderator
Internet Statistics - Top
|
|
|
