Everybody wants to increase revenues and many focus on getting more traffic to their site through Search Engine Optimization (SEO), Pay Per Click (PPC/SEM), Social Media Marketing, and more. However, there is one item that many, even seasoned, online marketers forget about — testing different versions of the same page or entirely different pages to maximize sales with your existing traffic. We all want to increase good targeted traffic to our sites. But you should also spend time making changes to your site to increase the conversion rates. Let’s assume that for one of the merchants you are promoting:

1. One visitor in 100 that comes to your website converts into a sale
2. Your average conversion value (commission) is $40
3. You are getting 10,000 visitors per month and as a result earning $4,000 from that traffic

Now how would those numbers look if you were to optimize your site and thus increase your conversion rate to 1 in 70 or even 1 in 50?

With these numbers, you can see the value in putting some effort into increasing your conversion rates. So how do you increase your conversion rates? You make small and large changes to your pages and website and test to measure the differences. These can be small changes like changing the text on the Order Button from “Buy Now” to “Order” to “Checkout.” These can also be large wholesale changes like a new site deign.

Most seasoned online marketers usually have a good feel for what will and will not work, but it is always a good idea to test. Let the numbers make the decision, not your gut feeling.A/B Testing
The “simplest” test is an A/B test. This is when you test two different versions, or variations, and measure the results. While most look at sales or conversions you can also look at things like newsletter signups, pages visited, time on site, contact forms filled out, phone calls, messages posted, etc. Any actionable event by a user can be tested.

Similar to the checkout example mentioned earlier you may want to test how to get more people to the merchant site you are promoting. So you can change the hyperlink from “More Info” to “Buy Now” next to your descriptive text and see which gets you more clicks.

You can also test a new site designs by redirecting some of your visitors to a new version of the website. I’ve seen 30% and even 40%+ increases in site redesign tests like this. There are a number of ways you can do this — the two most popular are to use an inner folder or a different sub-domain for the new design. This way if your new design negatively affects conversions, it will only affect your test subjects.

Multi-Variable Testing
As the name implies, Multi-Variable Testing (sometimes called Multi-Variant Testing) is similar to A/B testing with the addition that you are testing more than two changes. I will often perform multi-variable tests in my PPC campaigns and then implement the best producing combinations on the rest of my site.

A “simple” example would be changing the text on the order button and at the same time trying it in different designs:

As you can see in this test, we are testing the text on the order button combined with three different page designs.

You can really go crazy with this to get your optimal conversion rate. Last year I ran a test that had 42 different versions.

Timing
It is important to make sure you are comparing apples to apples when running these tests. Many people will run one version for a week, then another version the next week and so on. The problem is that you may have different traffic or different market forces at work.

Imagine a retailer that focuses on holiday gifts. The traffic during the first 2 weeks of December will be vastly different then the traffic during the first 2 weeks of January.

Ideally you will have a program/system that randomly sends visitors to your different versions as they enter your site in real time. Let’s say you have 2 versions and want to test it on 10% of your traffic — then your system should randomly send 90% to your regular page, 5% to version 1 and 5% to version 2.

You can build it so that these are different physical files or the same file with different variations.

You can even use ClickBank TIDs to measure the differences.

Tools
There are a number of options for setting up these tests:

• Build your own tools with a bit (or a lot) of programming
• You can use no-cost tools like Google Optimizer
• Paid tools and services like SiteSpect or Optimost

Each of these has different costs and each has its advantages and disadvantages. I suggest you do some research to find out which tool gives you the functionality you need at the price that works for you.

Summary
While we are all trying to get more relevant traffic to our sites, our goal is to generate more revenue. I suggest spending a portion of your time on increasing your website’s conversion rate, putting more money in your pocket.

Alon Cohen is the general manager for several successful ClickBank products. Check out the affiliate program for one of those products at: http://www.reversephonecheck.com/affiliates/

Note: This article was originally published in the Clickbank newsletter on October 30, 2009

When we query a single row (LIMIT 1) or a small amount (i.e. LIMIT 0, 30) it runs very fast. But when ever we do something that needs to run through most of the table (like a specific select that does not exist) it takes a very long time. Also group by functions like count run slow.

All of our queries will be through PHP pages.

Ultimately we want to create browesable a directory.

Part of the work my entail actually importing the raw data into the MySql database. We have it broken down into about 1200 separate insert files. Right we have about ½ of the data imported so far.

The final result of this project will be us being able to run fast queries against the database to create a directory similar to the one on the lookupanyone.com website mentioned above. Name specific queries (ie when we are searching for only a single record) should render in the PHP in under a second. GROUP queries  should not take more then 10 seconds per page.

Contact info:
Email alon [at] accucomcorp [dot] com
Phone (617) 262-9317
Fax (617) 262-9325

No recruiters please.