The Lost Art of Conversion Optimization

Commercial websites aren't cheap to build and maintain, and if you own a business with a website you probably want something to come from that expense. For some businesses that means a sale, for others a lead, and for yet others it's clicks on paid advertisements.

Regardless of what you want from your website, it seems obvious that more traffic to the site would help with all of the above...right? Yes and no.

Before you drive more traffic to a website, it is important to make sure that your website is doing what you want it to do as efficiently and effectively as possible. There's no point pouring more water into a leaky funnel. Enter conversion optimization.

If one were to define conversion optimization, the definition would be quite simple: making a website simpler and easier for customers to use.

Have you ever made a purchase online? Filled out an online form for more information? Signed up for an online service? If yes to any of these, how was that experience? Easy? Frustrating? Keep those experiences in mind as you read the following.

Though conversion optimization techniques can range from the ridiculously simple to the incredibly complex, the process can be broken down into 3 steps: Research, Measure and Test.

Research

One of the most beneficial things you can do, both for your business in general as well as your conversion rate is to exactly define and then thoroughly research your target audience. Only when you really know the people you are trying to reach can you craft a compelling message that will get them to do what you want them to do.

Visit successful websites in your industry, as well as successful websites in different industries that cater to the same demographic, and make notes about the site. Go through the following steps on each site:

1. Start at the home page, search for and locate a product you want, and add it to your cart (or start at the home page and find a contact form, if lead generation is your goal.) Make a note of how many steps were in that process, whether it was easy or hard to find the product/form, and anything else that stood out to you.

Also pay attention to how you feel and the colors being used for backgrounds, headers and buttons. Colors can evoke an emotional response, so keep that in mind.

2. Spend some time looking at 3-4 product pages. Are the product photos clear, attractive, and detailed? Can you zoom in or rotate the product image? Is the description clear and compelling? Is the price clearly displayed? Does the product page mention things like free shipping, discounted pricing, or anything else that would motivate you to make that purchase? Is what you need to do to buy that product obvious? Is a return policy and/or privacy policy visible? Make notes about all of these things on at least 3 product pages.

3.  Make a purchase. If you don't want to spend any money, you can stop at the point that credit card info is required, but it is highly recommended that you test the process from start to finish. Consider it an investment in market research. If you do lead generation, fill out the form with your real info (so you can see how they respond to potential customers later on.)

Do they force you to create an account to make a purchase? How many pages/steps are in the checkout process? Does it estimate your shipping costs? Can you ship to an address other than your billing address? Does it force you to duplicate information? Is it frustrating, or simple? Do you see security elements in place (VeriSign, TRUSTe, 128 bit encryption?) Are you comfortable making a purchase from the site?

4. See what happens after you make a purchase or fill out the form. Does the company contact you, and if so how quickly? By phone or email? Are they courteous and personable? Did the product arrive quickly, well packaged, as described?

By following these steps, you'll learn what is and isn't working for your competitors. Distill what you've learned into a list of things that you think were helping conversions. Keep in mind that at this step you are simply going to make a list of hypotheses, things you want to test on your site, so trust your gut. If any step in their process was difficult or made you uncomfortable, feel free to leave it off the list.

Measure

Google AnalyticsIf you don't have analytics in place, such as Google Analytics, then all your research and testing will have been in vain. You can't very well conduct a test without data!

Make sure that your analytics tracking code is properly installed on each and every page of your website. The quickest way to do this is by using ObservePoint. If analytics code is improperly implemented or missing from a page, visitor interaction with that page will not make it into your analytics data. That's a bad thing.

The most important thing you can do in your analytics program is to define and track your conversion funnels. Knowing the necessary steps to a conversion can help you to tighten the funnel and more effectively funnel visitors to the action you want them to take.

Monitor this funnel for pages that "leak", where visitors are leaving or doing something other than what you want them to be doing. These leaky spots are where you want to test conversion optimization changes, fixing the leaks so to speak.

Test

Google Website OptimizerNow that you have a list of things to test and the tools in place to measure, start testing them one by one on your website. The easiest way to do this is by using Google's Website Optimizer tool. This tool integrates with a wide variety of content management systems, and allows you to A/B test changes to your pages by showing 2 different versions, rotating with every other visitor.

In this way you can test your hypotheses one by one, determining with hard data which is the best way to do things. DO NOT GUESS. Test everything. When you've had at least 500 people view each version of the page, you have the minimum amount of impressions you need to declare a winner. More is better, but 500 is enough.

Once you have a winner, implement that change on the site and move on to the next hypothesis. Test, measure, test, measure, working your way down the list. Add to the list whenever you think of or discover a new possibility and test some more.

By researching your customers and your competitors, testing potential changes, and monitoring how people interact with your website, you'll be well on your way to increased conversions.

Guest Author: Sam McRoberts is the CEO of VUDU Marketing, a Utah SEO company. He has been involved with online marketing since 1999 and has worked with hundreds of clients, from small local businesses to Fortune 500 companies.

  • Tracy

    Great post Mirna...nice to have a refresh...I know sometimes I get stuck in the "traffic generation" wheel far too long...I've been spending some time tweaking autoresponders, optins...conversions...asking for the sale...all that good stuff. Thanks as usual...you rock!

    • Great job Tracy! That is where the time should be spent. I know it is time-consuming, but it is well worth the time. 🙂