February 2002

Traffic Generation Technique
When NOT to Drive Traffic
by Tom Antion

One of the biggest mistakes people make is to put Herculean efforts into driving traffic to their site before the site is thoroughly tested.

When I say tested I mean two things. 1. Mechanically tested i.e., does everything including the shopping system work smoothly? 2. Does the site convert visitors into buyers?

I had a lady start using Kick Start Cart http://www.kickstartcart.com  with a merchant account and a gateway. The moment she threw the settings in the cart she announced her new ebook to her large mailing list of fairly substantial people that she wanted to impress.

RESULTS: She got buried with complaint emails because everyone that tried to buy the book go their credit card declined which embarrassed them and embarrassed her.

It turns out that she had the wrong setting for the gateway to the merchant account and her AMEX merchant account had not yet been activated. I'm not sure exactly what else was wrong. I just know that a site that isn't thoroughly tested over a period of at least days is likely to cause trouble.

Test all the credit cards you accept in a live transaction. Get your friends to do it or you may want to sneak a few transactions through using your own credit cards (I'm not advocating this because you may violate your merchant agreement by using your own card in you own system.)

If you ask, most merchant account providers will give you a fake credit card number to use in testing. I have found though that there is nothing that compares to testing with a real credit card. The fake cards test your system, but don't prove that the money actually gets to your bank account which is the whole idea right?

TESTING YOUR SALES PAGES
Once all your links on your site work and go to the correct places and once your shopping system absolutely works, you must begin testing your sales pages.

If you send 1000 people to a sales page and no one buys, do you think it's a good idea to simply send more people to the page? Of course not.

The idea is to send a certain number of people to the page and track the number of people that buy.

Let's say you send 1000 people to your sales page using a pay-per-click search engine and 30 people buy your 19.95 audio tape. You have a 3 percent conversion rate calculated by dividing the number of people who bought by the number of people who visited the page.

This is pretty darn good. Now you change something about the sales page…Maybe you change the headline or maybe you change the call to action or maybe you change he entire page.

Your original page is now called the "control." You now test the changed page against the "control" and send another 1000 visitors. If you do less than 3 percent conversion on the new page, you go back to the control page. If you do better than 3 percent you now have a new page that makes you more money. It is now the "control" for future testing.

You can keep testing until you get tired or you get to the point where you can't get any higher conversion rate.

Now you can drive as much traffic as you want feeling confident that you are selling as much as possible from your sales page.

THE FINANCES
Let's say you end up with a page that gets 3.5 percent conversion on a $19.95 product and each product costs you $2.00 to produce and package.

So, for every 1000 people that hit the page you sell 35 products.

35 x $19.95 = $698.25 Gross Income

Your cost of goods sold (COGS) was $2.00 / product x 35 products sold = $70.00

Let's assume your pay-per-click search engine cost you .20 per click for your keyword to drive the 1000 visitors to the sales page. This would be .20 / visitor (click)  x 1000 visitors = $200.00

We'll assume the customer is paying the shipping and handling fee which should cover your cost of postage and labor to haul the products to the post office.

Here's the math:

$698.25 - $70.00 - $200.00 = $428.25 net profit

If you have a downloadable product you wipe out the COGS and shipping hassle  which adds an extra profit of $70.00 / thousand visitors. In the above example you would make $498.25 profit per thousand visitors.

If your pay-per-click keyword costs more than .20 that obviously would reduce your profits and if it cost less than .20 you would make more profit. If you get high rankings in regular search engines to drive part of the traffic and/or drive traffic from your ezine, media or speeches you would increase your profit even more.

We'll talk more about conversion rate in a future issue. But once you have your site thoroughly tested and have your best conversion rates calculated, you can drive as much traffic as you want.

Future Topics for this section

• Trade links -- it's great for you in several ways

• Write one articles and get at least five uses and maybe 50 uses

• Off line ways to drive traffic

• Online ways to drive traffic

• When NOT to drive traffic

Back to February 2002 Index