Fact and Fiction: Boosting Your Conversion Rate

Go to article

By Jenny Elliott

One of the most common questions posed to web analysts is, “Why is my conversion rate only X.Y%?” This question simultaneously makes me clap my hands with joy and cringe. It means some real detective work is ahead, but all too often the powers that be do not want to do the work required to find the answer.

 

Why aren’t more people buying? Usually it is a combination of smaller missed opportunities over the course of a week or a month adding up to a lot of lost conversions. Executives want there to be one answer - one area where change needs to happen. While there is rarely such a simple solution, there are a few common areas to investigate to get you started.

 

Examine your checkout steps

Review the checkout funnel flow for two things: exit points and backwards flow. Find out what percentage of people checking out proceed through linearly without any back and forth. If people do take a step back, what is the most common path? Examine what percentage of people visit a step in the checkout process more than once during their session. This can indicate customer confusion. Often messaging is the culprit; the customer expects to see some information on the page but does not.

 

Once the leaky points are identified, perform qualitative user testing to truly find out what customers are thinking. Although metrics can show us where shoppers abandon, this is not necessarily the same place where they get frustrated. When it comes to the cart, knowing exactly what you need to fix and how to address the problem requires direct feedback from your customers.

 

One of the best ways to convince the boss to spring for some usability testing is to illustrate what the conversion rate would look like if you stopped 50 percent of the people currently jumping ship. Times that number by AOV and show the potential sales that could be won.

 

Segment conversion rate by channel

Overall site conversion is reviewed frequently, but it does not come close to the whole story. Review conversion rate by channels such as PPC, SEO and email marketing visitors. Then dive deeper into lower performing segments - perhaps the culprits are some severely underperforming marketing campaigns that are dragging the overall site average down.

 

I like to re-calculate the conversion rate minus the segments/campaigns I believe could be the problem. You must show the higher-ups what could be if given time to identify exactly what needs to be corrected.

 

High bounce points

When evaluating visitor segment conversion rates, start with the easy stuff - the pages that are high bounce points for those specific users. Then work backwards. What keyword did they use to get there? What did the PPC ad that drove them there actually say? Part of understanding why a visitor did not stay on your site is knowing the intent that drove them there. This can lead to content and layout updates so visitors more clearly know what they are getting what they are looking for. A/B testing is a great tool to test out different landing pages and find out exactly how the site needs to be improved.

 

Non-transactional conversions

Users come to a site for a variety of reasons, including store hours, check return policy, check order status, write reviews - the list goes on and on.

 

With a little work, you can identify and segment these visitors out of the general population and measure their respective conversion rates according to what they were looking to achieve on the site. Try recalculating your overall site conversion rate with these people taken out, as their intent was not to buy when they hit the site anyway. Not only will you get a more accurate representation of the potential buyers but you will gain greater insight into how well you are doing with these other tasks. These interactions are important. Making the store locator difficult to use or having convoluted return policy information will ultimately affect a visitor’s decision on whether they want to buy something from you in the future.

 

Get real

Some sites will give customers the option of buying an item in-store and provide links to other sites to purchase from. Or perhaps the product mix is such that not all items displayed are buyable online. In both these cases, I see low conversion rates. But, that is to be expected when you are helping people out the door and displaying items they cannot buy. Businesses need not altogether stop these practices, but instead may want to get realistic about how this practice affects conversion rates.

 

My recommendation for identifying your desired conversion rate is to take the time to gain confidence in site usability, ensuring traffic-driving methods are finely tuned. Whatever that conversion rate is at the end of those exercises is probably what you should count on and measure against as you go forward and grow the business