What is Google Optimize?
Google Optimize is a Google AnalyticsA/B testing solution. It is a relatively new A/B testing tool – but it’s a Google A/B testing tool so the expectations are high. It was globally launched out of beta just last March. The tool allows businesses to experiment, test, and personalize variations of a page or even two different pages of a website (we’ll discuss this later on in the article).
One of Google Optimize‘s key selling points is the native integration with Google Analytics. This removes the necessity for additional tools by addressing different marketing and conversion channels. Other capabilities and features of note are:
- Three experiment types: A/B Tests, Multivariate Tests, and Redirect Tests.
- An experiment management features. These built-in features include an activity log, experiment preview, and user permissions
Advantages of Google Optimize :
- Easy test setup and deployment. This is the result of the WYSIWYG visual editor. The editor removes the need for a tech team, especially for the creation and deployment processes.
- Native statistical reporting. The native integration with Google Analytics means that there is little to no conflicting results and that data analysis is straightforward and highly relevant. You can test against KPIs without any additional configurations and access the results both in Google Analytics and Google Optimize.
- Advanced Audience targeting. As a Google Analytics A/B testing solution, Optimize has the capability to target visitors based on their behavior, interaction with your website, and location. The capability to target predefined Google Analytics audience is only available in Optimize 360 though.
Google Analytics A/B Testing: Free or Paid Version :
The private beta version for Google 360 kicked off in March 2016. The beta was only opened to the public after a free version was announced six months after. According to anA/B testing tools comparison by Convert, Google Optimize 360 can cost more than $150,000 USD per year. It’s quite expensive compared to competitors like Optimizely whose Enterprise edition costs only about $35,000 per year. But surely there’s a reason for Google Optimize 360’s high price? Let’s take a look at differences between the free and paid versions.
It’s expected that the free google optimize version would have limited features and capabilities compared to the paid version. Let’s look closer at these differences:
- The first difference is with audience targeting. In the free version, there are only three target metrics to choose from while the paid version allows the full targeting options available in Google Analytics.
- You can only test 16 variations in multivariate testing.
- There are only 3 experiment objectives you can choose from in Google Optimize while you can use any objectives available in Google Analytics goals. You can even change the experiment’s objective to test how the results vary from objective to objective.
- You can only conduct 3 experiments simultaneously.
- User and account administration
- Implementation services
- Support and services
With Google Optimize 360’s price alone, you can get a picture of how large and sophisticated the businesses Google is trying to target. There’s an assurance of quality with GoogleA/B testing though.
Set Up Google Optimize A/B Testing Tool :
1. After learning the details of how Google Optimize works, it’s now time to try it out. First, you need to sign up. In order to experience a seamless integration, sign up using the same Google account that you use for Google Analytics.
Aside from this, there are also other prerequisites in the Google Optimize setup process:
- Installed Google Analytics web tracking on your website.
- Chrome web browser.
- Google Optimize Chrome extension.
Once you are done, it’s time to setup your account in order to start your Google Analytics A/B testing experience. But at this stage, we’ll only touch up on the necessary set up. There are parts of the Google A/B testing process that aren’t as straightforward for beginners so we need to break down each step.
2. Create an account and container :
This account refers to the representation of your business or company where all experiments are done under. It is so to say “the highest level”in the hierarchy. You can use your business or domain name as the account name.
You can setup multiple accounts in Google Optimize but only one account for every company or domain. You can also manage user access for every account.
The container is the repository for all configuration information of experiments in your account. You can either set up a container for a single domain in an experiment, or a container for multiple domains in an experiment. You can base structure of your containers based on user journey or your test objective. Well-organized containers enable you to have a more seamless Google Analytics A/B testing experience.
Once you are done, you will be directed to the Experiments page.
3. Link Google Analytics :
There are two places where you can find the option to link to Google Analytics: from the onboarding checklist or from the main menu. On the right-hand side of the Experiments tab, you will see an onboarding checklist with drop down arrows. You can see that the “Manage accounts and users” is already ticked. Skip the “Create an experiment” for now and expand the “Link to Google Analytics”.
Click the “Link Property” button. A pop-up will appear prompting you to enter a Google Analytics property from a dropdown list. Select the property and one of the two Analytics view options. Finally, click the “Link” button.
If you want to add or edit a property in the future, just go to the main menu and select “Accounts”. Select the container you intend to use. The container information will then appear.
Next to “Google Analytics” is an edit icon. Click the icon to edit the property to the one you want to link to. Edit the property from a dropdown list and then select a view and then click “Save”.
You can only link one property to a container but you can link the property to up to 10 containers.
4. Install Google Optimize Snippet :
Next on the checklist is the installation of the Optimize snippet to your website. The main snippet is added to the tracking code already on your website. This snippet loads a container as a Google Analytics plugin. Note that you need to have basic knowledge on editing HTML code for this.
First, expand the “Install Optimize snippet” panel and then click on the “View Snippet” button. As you can see on the tidbit of information below at the bottom of the panel, you can install the snippet through either Google Analytics or Google Tag Manager. It is recommended to directly add the snippet code to your website code. But if you want to do it with Google Tag Manager, you need to read this guide even if you already know how to use GTM. It’s explicitly stated that the modified Analytics tracking code should still be installed directly on your website and not through GTM.
But back to the installation of the snippet, just follow the instructions in the dialog box that will appear after clicking the “View Snippet”button. One obvious disadvantage of this method compared to using the GTM is that it can become time-consuming especially if you have several pages you want to include in your experiment.
There is also an optional snippet you can also install, the “page-hiding snippet”. The page-hiding snippet’s main function is to prevent page flicker. Page flicker happens when a user of page visitor sees visible changes on a page as the experiment variants are loading. The page-hiding snippet prevents this from happening by hiding your page as the container is loading. The snippet is added to page’s code just after declarations. For more information on this subject, check this guide on the page-hiding snippet deployment.
5. Install Google Optimize Chrome Extension :
This obviously requires the Google Chrome browser. Without this extension, you cannot run the visual editor which is necessary for the creation of variants and other Google Analytics A/B testing tasks.
Setting Up the Google Optimize A/B Testing Experiment :
The time has finally come. But before anything else, make sure that you have a hypothesis to test.
At the end of the experiment, the hypothesis can be proved or disproved depending on the accumulated data.
Now, to begin the experiment, make sure that you are on the experiment page. Click the blue button that reads “Create experiment”.
A dialog box will then appear. Fill in information like experiment name (255 character limit), the URL of the page to test (editor page), and the type of experiment.
If you are new to webpage testing or you are only familiar with A/B tests, read this overview of the experiment types. The illustration below.
If you are unsure on what type of experiment to run, check this article by an A/B testing company about experiment types and when to use them. For this article, we’ll be creating an A/B test since it’s the simplest test to run. Note that Google Analytics A/B testing and other A/B testing companies may use different terminologies on the different types of experiments but they are all structurally the same.
Once you select the A/B test, you can either configure the objectives and target or create a variant first. We’ll do the configuration first so that we can get it out of the way.
Necessary Configurations before A/B Testing with Google Optimize :
Google optimize Objectives :
- Select an analytics view.
- Select the Primary Objective. There are three pre-selected objectives so you can click “Add an Objective” to add either one or both of the remaining objectives as secondary objectives. You cannot change your objectives afterward so make sure that you selected the right and relevant objectives.
- Add a description for the experiment. This is also where you can add the hypothesis you constructed at the beginning of the experiment setup.
- When you are done, click “Save”.
In Google Optimize, you have several options in targeting. You can either use geotargeting, behavior targeting, technology targeting, and audience targeting. In order to fine tune your targeting, Optimize has two targeting controls: Who and When.
Decide on the percentage of visitors and weight of visitors to target. The visitors’ “weight” pertains to the probability of a visitor being shown one variant over the other. The weight of the variants is equal by default (50/50) wherein each variant is shown every other visitor.
If you have more than two variants or want to show a variant more, click “Edit” to adjust each variant’s weight. The weight percentage should all add up to 100.
When to target :
The “When” targeting control determines the timing of the variants’ showing. You can set targeting rules or conditions to determine when and where a variant shows. So every time there is an eligible visitor to page, there is an evaluation whether the visitor met the requirements that allow them to see a specific variant.
Take note though that this won’t affect a visitor’s eligibility in participating the experiment, just the variant they will be shown. Note also that though the Google Analytics A/B testing targeting is similar to Google Analytics’ targeting options, there are differences and limitations especially on the free version.
Targeting rules :
In order to set targeting rules, you should create one in the condition builder. You can simply click the “+CREATE RULE” under the “Rules for when to target” in the “Targeting” tab. You can also edit and delete an existing rule. For more information on how to add rules, read this guide.
You can combine rules in order to create a very specific rule that matches your needs. Just remember that there are different targeting rule types you can build these rules upon.
Now that you are done configuring the objectives and targeting rules, it’s time to create a variant and then run your experiment.
Variant Creation and Editing in the Visual Editor :
The creation of a variant is pretty straightforward. You can add a new variant from the experiments view. Select the variants tab and then click “+NEW VARIANT”.
But in order to edit the variant, you need to head to Optimize’s visual editor in Chrome. Familiarize yourself with the visual editor interface since you will be using it a lot. Here are the main components of the visual editor:
- App bar
- Editor palette
- Current selection
To learn more about the visual editor’s components and tools, read this thorough introduction from Google Analytics A/B testing support page.
Going back to variant creation, click on any page element that you want to edit. In the editor panel, make the changes you want (e.g. change text content). Click “Save” and then click “Done”. The end result looks something like this:
Once you are done editing, click “START EXPERIMENT”. Check if the status field says that the experiment is “Running” to make sure that the experiment is already live.
Reporting and Data Analysis
The Reporting tab is located at the top of the experiments page. You can monitor a running experiment or access the results of a finished experiment from there. These data can also be accessed from Google Analytics. The Optimize reporting tab has three main elements:
Summary Card :
Displays the experiment’s status and the summary of the results based on the experiment’s primary objective.
Improvement overview card
Compares the performance of variants in comparison to the control (original) variant.
Objective detail card :
The third card in the reporting tab displays the performance of the experiment’s variants against the set objectives.
It is recommended that the experiment lasts for about two weeks or until one of the variants reach 95% probability of winning. That’s why it’s important to wait for more data especially in the objective detail card to have an accurate insight on the experiment’s result.
Now it’s your turn , start implementing Google optimize into your buisiness. And if you liked this tutorial do not forget to share it 🙂
- Adplexity Coupon Code $80 Off and Walkthrough - September 23, 2019
- SiteGround vs GoDaddy: WordPress Hosting Comparison - September 13, 2019
- SiteGround vs Inmotion: WordPress Hosting Benchmark - August 31, 2019