[Alaister]

What conversion rate optimization strategies have you seen work the best?

At Freelancer.com we do a lot of A/B testing of different strategies and hypothesis to maximize conversions. We've found that showcasing beautiful examples of work created by our freelancers and having strong social proof increases our conversions.

[yoangov]

For me there are 2 main factors that are the most important for my conversion optimization job.

1. Simplicity.

Everything should be super easy to be accessed, and done in the most straight forward way. Sometimes the design outcome of this could be uglier, but prospects will convert better. I believe that websites should be as simple to use as possible. People are building huge mall centres in order to attract visitors, which in offline marketing is true.

However, in the online field you are dealing with technology, and people get lost, as they can't deal with computers. If your design is too shiny, and customers have to click more than 3 buttons before they place an order, how many could get lost out of 10,000 visitors?

Exactly.

2. Statistics.

Statistics is the other highly important factor. It can make you or break you, and a lot of other marketers are doing it wrong. They run A/B tests and they have no idea how many clicks exactly they need before they can decide whether the test is successful or not.

In order to do that you have to use and calculate the Statistical Significance. If you don;'t get that factor most of the time you have 1/3 chances to make a mistake on which is the winning concept. Then you need to be able to correctly measure the statistical power from the test. Then, and only then you can make correct decisions, based on true statistics.

This usually requires more patience, and higher budget, but this is the way to optimize your stuff the correct way. A/B testing is gold, but not many people know how deep it can go, and way to make the chance for mistake 1 out of 20+

I'd love to see a nice advanced discussion in this board

[geegel]

A good first step is to realize that not all traffic is created equal, so each traffic source should be treated differently with a separate landing page.

In essence what I'm talking about is segmentation. Take the time to understand and cater to each segment of your audience and your conversion rate will dramatically increase.

Regards,
George

[richpage75]

Unfortunately there is no 'silver bullet' that will work well with every website.

Here are some important step to creating tests with better impact though:

- Always get voice of visitor feedback to create test idea insights and hypothesis - use tools like usertesting.com to ask visitors for their feedback - what they like, don't like, etc.

- You should also use tools like clicktale.com to see recordings of what people are/aren't doing on your website. Very eye opening and great for forming test ideas.

- Use your web analytics to find out which pages are most in need of improving. Look for entry pages with high bounce rate, and your top exit pages. Always check your drop off rate for your checkout or signup flow too. Then optimize these first - this will quickly give you biggest impact.

I could write much more - you should consider checking out my free toolbox for more optimization ideas. CRO is much harder than most people think

[pnglifesolutions]

richpage75, those are some interesting tools you mentioned. they look costly though.

I shall try the web analytics route first.