AB testing is important, and can increase your app's performance. This post shows a quick example of how to perform AB testing using Google Analytics and GA API.
Note that this article assumes you have set up Google Analytics in the first place (it will only work with Google Analytics).
If you want to learn how to do AB testing without using Google Analytics, check out this tutorial.
What is AB testing?
AB testing is a technique used to measure the relative performance of two apps or pages on mobile devices. In simple terms, it measures how fast one page loads compared to another.
It works by loading two different versions of an app into the same user's device. Then, if the user interacts with them both at the same time, we see whether the faster version of the app loads faster than the slower version.
The most common way to get started with AB testing is to compare the performance of two static pages (like a Google search bar and an image gallery). To do this, you will need to install a website like: https://www.google.com/analytics/webmasters/static-pages/home.html,xe2x86xc6c5vhgqdxt9sqj7fmt0qy.
A good example of what I am talking about here is the blog post I did last week in which I showed how to test the load time of the Google Search bar vs. the Android browser.
Google Analytics offers a free web analytics tool that can be used to perform AB testing for your own site. You don't have to pay anything up front – just sign up for a free trial!
Let's take a look at how to create a test plan using this tool…
1. Set up your test account
Head over to the "Analytics > Google Analytics > Setup" page and you will need to fill out some basic details, like your email address and your Google account name.
By clicking "Create a free account" you will be directed to the Google Analytics login page. Enter your email address and click the "Create a free account" button.
Now it's time to set up your test data. For this example we'll use a test account with no search terms or traffic sources as we'll see later on why such a test would be useful.
2. Create a new test plan
To do this, head over to the Google Analytics dashboard and click the "New Test Plan" button.
You will then be prompted to enter an email address where you want to receive notification when your test runs are completed.
3. Create a test campaign
This is where you select the type of test you want to run. In this case, we will have our test users browse to our homepage for one minute at a time.
Once you create that test profile, you can proceed to the next step.
4. Choose the type of test (a) Classic or (b) Custom
There are two different ways to choose what type of test: You can either manually specify the URL path in which visitors go to before they arrive at your website, or you can also let your web analytics tool automatically determine the path based on your website's structure.
For example, if you had a blog-like website, you could generate a URL like stitch.me/blog by entering www.stitch.me/blog into your browser's address bar.
If you were using AdWords, you could send visitors to a landing page associated with your ad group instead of the actual blog post itself.
5. Choose the number of visitors per session
Here you must input the number of visitors that will visit your site each day. For example, if you have multiple blogs and one of them has 10,000 visitors a day, you would enter 10,000 as the number of visitors per session.
6. Enter the duration of the session
In this step, you must enter the amount of time that the visitor stays on your website. The duration is divided into three parts: minutes, hours and days.