What is Google Optimize?

What if I told you that it could predict your test’s success before you’ve even implemented it? Or, that there’s a tool that gives you suggestions on what test you should run?

A conversion rate optimization (CRO) dream.

Half the struggle of testing for better performance is figuring what to test and what exactly the results are telling you to do.

Google Optimize gives me all the heart eyes. It is one of the most common, easy-to-use testing tools. It eliminates the struggles of testing, so you can optimize better.

In this article, we’ll talk all about the basics of Google Optimize, including the experiments you can run and how to get set up.

Google Optimize is a web optimization and testing tool created by Google to help increase CRO for your websites and landing pages.

It allows you to test variants of web pages, similar to other testing platforms that allow you to test the functionality of your website, along with your CRO and user experience (UX) hypotheses.

Google offers both a free version, named Optimize, and a paid version named Optimize 360. 

Optimize is made for small to medium-sized businesses getting started with experimentation. We recommend using Optimize if you’re not looking to perform a multitude of tests.

Optimize 360, on the other hand, is best for large enterprises with more complex testing needs.

You can read more about the features of both Optimize and Optimize 360 here.

Optimize has an edge over competitors for many reasons, as it allows for easier setup, enables more advanced targeting and reporting, and empowers quicker implementation of what you’ve learned.

But above all, Optimize’s seamless integration with Google Analytics (GA) is what truly sets it apart from other tools.

What are the features of Google Optimize?
While there are many features to Optimize, we’ll be touching on the most important ones including

• experiment types
• visual editor
• objective types
• and reporting

Optimize is similar to other A/B testing tools, in that it allows you to run A/B testing, A/B/n testing, multivariate testing, and split testing.

However, it also offers additional types of experiments, including redirect tests, server-side experiments, additional simultaneous experiments, and personalization.

Redirect tests, also known as split URL tests, are so similar to A/B tests, they could be considered a type of A/B test.

With these tests, you can test two versions of the same page or two completely different pages.

With server-side experiments, you can create and deploy variants on your system, while using Optimize to view reports and determine winning and losing tests.

The experiment code is directly inputted from the server-side, which means the code is set up to show the best variant option to users automatically. 

This experiment type is mostly used for more complex changes.

This experiment type is exactly what it sounds like: simultaneous tests so you can get your results more efficiently. You can run multiple experiments at the same time.

With Optimize, you can only run up to 5 tests simultaneously.

With Optimize 360, you can run up to 100 experiments simultaneously.

Personalization in Optimize allows you to customize info on the page to be personalized to the user by utilizing data you have such as geographical location, age, device, and the like.

You may have been subject to personalized targeting like this if you’ve seen “products you may like” or similar offers promoted to you on websites.

The algorithm aims to make sure you have the optimal opportunity to fulfill your users’ needs and help you increase your CRO.

This What You See Is What You Get (WYSIWYG) editor allows you to create a new variant for experiments without worrying about re-coding your site for each test. 

You also have the option to edit the raw HTML or add JavaScript or CSS rules directly into the editor if you choose.

When you create an experiment, you edit a copy of the webpage, and then this redesign will compete against the original once testing starts. It can also reconfigure your experiments for different device sizes, which can be especially useful for responsive design.

But before testing starts, this visual editor will alert you of any potential problems it foresees in your experiment. This feature allows you to uniquely optimize experiments by saving you time and setting you up for success from the beginning.

It doesn’t make much sense to test something if you don’t have goals, or an objective, to compare your results against. 

Thankfully, Optimize and Optimize 360 provide several experiment objective types for you to build your experiments on.

System objectives are the most common type of objective used for experiments. These objectives can be page views, bounces, and more, and are already set up for you to add.

Custom objectives are objectives that you create. You can build objectives specific to your goals in Optimize. Custom objectives are a good option when System objectives aren't sufficient to measure your experiment.

If you choose to, you can also set your Google Analytics Goals as experiment objectives. It’s an easy way to get objectives without having to set them up. If you’ve already set up events or goals in Analytics they will also be available as long as Analytics is linked to AdSense.

Additionally, you can use Upgraded Experiment Objects if you have Optimize 360, which allows you to set additional experiment objectives after your tests have started.

Optimize has a reporting option that allows you to view your experiment results. This is a critical feature since it can help you immensely with your CRO testing. If you can accurately and easily view experiment results, then you won’t know what to change that will be effective.

Optimize can also analyze the reports to give you suggestions for your experiments. However, we suggest you still take a look at the report yourself to see what you can glean, even if it’s your first experiment. 

Once you get through onboarding, you’ll be taken to the “Experiments” view with a default container named “My Container.”

You’ll find the account and container IDs by clicking “My Container” in the top left. You’ll need these IDs for the last step.

Now, let’s connect your Google Analytics account.

As we discussed before, Optimize’s integration with Google Analytics (GA) allows you to have a superior optimization experience, so let’s take time to link your GA account to your Google Optimize account.

If you don’t already have a GA account, check out our blog to get set up.

Go to “Settings” in the top right and click “Link to Analytics.”

Then, select a “Property” and click “Link” to link your account.

Finally, install the Optimize snippet on your website.

Go back to “Settings” and locate the Optimize snippet below “Link to Analytics.”

You’ll have two options: manually update each page with the snippet or use Google Tag Manager (GTM).

The easiest way to install the snippet is with GTM. If you’re new to GTM or want to learn more tips and tricks, you can check out our GTM hacks and best practices.

To start, make a new tag. You’ll see that Optimize is already on the list as a tag type.

Then, you’ll want to enter your Optimize and GA IDs that we located before.

Now, you can choose what pages you want to add this snippet to. These can be individual pages or your entire site, “All Pages.”

Remember to preview your work. Then save it, you’re done.

If you run into any issues with GTM or GA, you can use the Chrome extension Tag Assistant Legacy (by Google) to help troubleshoot. 

If you’re opting to install the snippet manually, you’ll need to add the code to the <header> tag on every page you want to optimize. 

We recommend working with a developer for this.

You’ve made it. We’ve talked about a lot of the ins and outs of Optimize, but we’re so excited for you to try it out for yourself.

Not sure what to test out first? Here are the A/B tests we prioritize when we’re looking to boost conversions.

Don’t stop there. You can test out form questions,  lead magnet ideas, social proof location, calls-to-action, and even landing page headlines.

Let us know how it goes.