According to Wikipedia, conversion optimization, or conversion rate optimization (CRO) “is a system for increasing the percentage of visitors to a website that convert into customers, or more generally, take any desired action on a webpage”. If you have an eCommerce business then this is pretty much your life’s blood. The issue becomes, not so much that you need it but rather how do you achieve it.

As you probably know, eCommerce websites have a bunch of moving parts. There are product pages, category pages, shopping carts and checkout pages just to name a few. Then there are the conversions (a sale) and micro conversions (signing up for a newsletter). This process, called a sales funnel, leads the visitor to a conversion. At any point while travelling through that funnel, the visitor can bail on you.

A visitor interacts with a site at many different levels. For example, there are clicks and mouse movements but there are also emotional interactions as well such as, happiness or frustration. These types of interactions are hard to evaluate with raw numbers. This is where conversion optimization starts.

There can be many reasons why a visitor abandons your sales funnel. Many of those abandonments can be explained. They look around and don’t see anything they like or perhaps they are not ready to buy. Those aren’t the ones that you need to worry about, at least initially. The ones that you need to worry about are the ones that go deeper into the funnel. Perhaps they put something in the cart but never checked out or perhaps they started the checkout process but for some reason abandoned it. Those are real issues that need to be addressed because they made the decision to start down the “funnel” but stopped for some reason along the way.

Part of conversion optimization is to hypothesize and test. Maybe the shipping is too high or they don’t like PayPal or maybe they were confused. How would you know? The first thing you need to do is to make sure the process is mechanically sound. You track mouse movements and mouse clicks to see if the design is not the issue. Assuming that the tunnel is mechanically sound, you can move on to emotional issues. Like messaging, pricing and payments options. This part is trickier. However, planning out a strategy to test these hypotheses is key to getting to the root of the problem.

You start with testing one option versus another which is called an A/B test. You can test different variations of the same issue which is called a multivariate test, but you don’t want to change payment methods and shipping at the same time because you won’t know for sure which is the reason for success or failure. The idea is to see if one leads to more conversions than the other then take the winner and test it against another variation to see if you can provide more conversions. Eventually you will find which variation is the right one.

The whole exercise is to make sure there are no roadblocks preventing visitors from completing a transaction. Increasing your conversion rate by one or two percent could mean hundreds if not thousands of dollars in revenue. However, you shouldn’t stop there.

The process should be an ongoing process continually tweaking more and more to squeeze out more and more percentages of increases. Do you ever reach the end? Yes and no. Yes, you eventually figure out the right combination or at least a rate of return you can live with and no, because a website should always be in a state of change. With new features comes more testing and the process starts over again.

Is this really necessary? If you have no real concern about the number of conversions on your site then you probably won’t see the value in this process. However for those who see their website as a revenue source then the return on their investment will be a no brainer. Trust me you will notice the difference.