How to create experiments in google analytics

UI or user interface is getting a much larger attention than ever before. Getting your webpages just right for your users is really important but quite a difficult task when you are designing something in the way you think looks best.Google Analytics offers a way for you to test different layouts for particular pages and then find out which pages are most likely to complete conversions than others. I want to show you in this post how you can also do this to improve your websites UI.

Step 1: Create a Google Analytics Account

[image]
http://www.jawilsondesign.com/wp-content/uploads/2013/03/signup.jpg
[/image]

If you haven’t already the first thing you need to do is create a google analytics account. This is a totally free thing to do and it will allow you to study your website, what is working, how users navigate your site, what pages are most used and which pages aren’t so successful amongst many other things. It is well worth your time and I am sure many of you already have this great feature on your site. If you don’t then what are you waiting for? Simply follow the instructions that google provides and sign up right away! ( If you need any help or guidance please leave a comment below).

[image]

http://www.jawilsondesign.com/wp-content/uploads/2013/03/setgoal.jpg

[/image]

Setting Goals

Goals are a versatile way to measure how well your site or app fulfills your objectives. You can set up individual Goals to track discrete actions, such as transactions with a minimum purchase amount or the amount of time spent on a screen.

1.Firstly on the main dashboard of analytics press conversions and a drop down list will appear like in the image above. Once about press goals > overview.

2. You will then be given a page like that below. Put in the url of the page you are wanting to set up an experiment in the blank box.

3.You can set up a goal for many things. How about making sure take the action of arriving on a particular page that will lead to a purchase? Use the url destination radio button. Or what if you just want people to stay longer on a page to make sure they read the information you are providing? Then choose Visit duration. It is pretty self explanatory. Once you have made your selection of which goal radio you will be given another form to fill in. For this example i am seeing how many people will arrive at my blog from the home page so using URL destination (See image below).

[image]
http://www.jawilsondesign.com/wp-content/uploads/2013/03/goalset.jpg
[/image]

4. Once you are happy with everything press save.

5. Your Goal is now active.

Step 3:  Access the Experiments Feature

With our goal set up, we can now start to create experiments in google analytics to determine which of our pages will perform the best in terms of visiting the blog from the homepage.
[image]
http://www.jawilsondesign.com/wp-content/uploads/2013/03/experiments.jpg
[/image]

1.Under the ‘Content’ section of the main left menu, select ‘Experiments’.

2. You will need to make several variations of your webpage that you want to improve conversions on ( i recommend at least 3 ) in order to have a more meaningful test. Whatever CMS you are using you just need to create more pages and publish them in order for google analytics to be able to find those pages you want to test the original to.

3. Watch this video from google and it will guide you through the rest of the setup.

4. Once you have completed this stage it just a matter of time waiting for visitors to come and try out your new page.
5. Google Analytics will give you a tone of results and a percentage of what the likely hood is that your new designs will outperform the other or not.

Step 8: Arriving at a Winner

As mentioned in the previous step it is just waiting for enough visitors to use the page to arrive at a winner. Whether it takes three days or three weeks to arrive at a conclusion, you can be confident beyond reasonable doubt that the winner of the experiment will indeed be the best performer in terms of the goal that is being measured.

Step 9: We Have a Winner, What Now?

Depending how confident you are that one of your new design will outperform your original design in terms of the conversion set then you can replace that page or not.

Also, you need to consider all variables that may have effected the experiment. How many people visited the original design compared to the other design samples. Do the visitors have the same tastes in design or not? As you can see it is important to analysis the data and not just take it at face value.
[line]
If you need any help with your experiment don’t hesitate to contact me and I will see what I can do to help you, best of luck in improving your UI.