We conducted 2 separate demo download tests. Each of these tests brought an increase in demo downloads and the overall net conversion impact was 19.66%

Test #1: What we tested
In this test we were faced with the challenge of getting visitors emails prior to any demo download. Prior our engagement, visitors who wanted to download a demo of the software had to register as new clients and were led directly to a login/registration page.

We saw bounce rates of over 50% on this registration page so we tried to come up with another way of getting an email address from the visitors prior to any download. We decided to test using an overlay pop up window.

Control

Variation 1

Test background
The client sells downloadable exam software for the IT industry.
-    test duration over 3 weeks
-    130,379 visitors took part in this test, generating 1,898 demo downloads
-    2 variations in the test

Test results
-    the demo pop up variation improved conversion and increased demo downloads by 10.49% at 99% confidence level

After this initial success, we still believed that we can do better and improve conversion even more so we ran another test, this time testing different pop up windows.

Test #2: What we tested
We put into the test 3 different demo pop up windows. We wanted to know if a smaller, more simplisctic version of the same pop up would bring any improvement.

This time, we created 3 variations:

Control

Variation 1

Variation 2

Test background
-    test duration 4 weeks
-    76,767 visitors took part in this test, generating 2,768 demo downloads
-    3 variations in the test

Test results
-    the demo pop up variation 1 improved conversion and increased demo downloads by another 8.32% at 96% confidence level

Overall, we achieved from 2 different tests, a net compound gain of 19.66%.

Calculation of the compound gain:
= [(1)*(1+T1%)*(1+T2%)]-1
= [(1)*(1+0.1049%)*(1+0.0083%)]-1
= 1 * 1.1049 * 1.083 – 1
= 0.1966 * 100%
= 19.66% (net gain)

Hypothesis
Here is some speculation on the possible causes of the test outcome.

• Test #1: Visitors don’t like registering and see the registration process as a hurdle. By making it look less like as a registration process, we were able to persuade more visitors to give us their email address in return for the demo download.
• Test #2: It seems like there is a fine balance between providing too much and providing too little. Too much information arguably made visitors feel a bit overwhelmed at the size of the pop up and this is why the larger pop up with all the information, failed. The same applies to the stripped down version of the pop up which didn’t contain any benefits in the bullet points, thus not giving enough incentive for visitors to carry on with the download process. However bullet points with the demo benefits were important during the decision process and this is probably why test variation 1 won. Sometimes less is more, but too little can hurt as well, which is probably why the smaller pop up version without the benefits in bullet points didn’t win either.

Would you like to increase your conversion rates and revenue? Contact proimpact7.com regarding conversion optimization at call +44 (0)20 7617 7848 or contact Jan Petrovic at jan@proimpact7.com.

If you are looking for a web analyst or analytics manager visit www.dashboardinspector.com and find out how much you can save by using both of our services (conversion optimization and web marketing analysis.)