Google’s Optimize 360 creates exciting opportunities for testing & personalization

By now, you’ve probably heard a bit about Google’s new Analytics 360 Suite of products, designed to give enterprises a truly integrated package of marketing analytics and optimization tools.

While some of the 360 Suite is composed of products that have been around for a while — such as Analytics 360 (formerly known as “Premium”) and Tag Manager 360 — several of the Suite’s products are brand-new. These include Audience Center 360, Data Studio 360 and the tool I’m most excited about: Optimize 360.

Optimize 360 allows you to run A/B or multivariate tests, as well as personalization campaigns, on your site. Of course, a variety of tools already offer these features: Optimizely, Monetate, Maxymiser and a host of others. So you may be wondering why you ought to be excited about the latest entrant in a crowded field.

If you’re already a Google Analytics (GA) user, Optimize 360 should be very exciting indeed. Here’s why: As a marketer, you can define important, valuable audiences in Google Analytics, then seamlessly port those audiences into Optimize, where you can test new, personalized experiences.

It’s a simple idea, but a key differentiator in the testing/personalization space. On top of that, you’ll also be able to run deep analysis on your test results, using Google Analytics tools you already know and love.

Let’s take a look at a handful of ideas for audiences that e-commerce businesses can use to get off to a fast start with Google Optimize. Then, we’ll take a look at why analyzing test results using the full power of GA is such a big deal.

Anonymous users

Anonymous users might seem like an odd target for personalization at first, but consider: If you want to reach potential customers with laser-targeted marketing campaigns, it helps to know a bit about them. There are many ways you can experiment with encouraging anonymous users to create an account, and hence a record, in your CRM (customer relationship management) system.

For example, if an anonymous user is browsing a product detail page, you could test a modal window that offers the user a promo code for their next purchase if they sign up to receive your email newsletter. Even if they don’t end up buying something during that visit, you might now be able to target that user with email marketing campaigns.

Alternatively, if you allow users to check out as guests, you could test a message that offers a price discount on the items in the user’s cart if they decide to create an account instead. The small hit you take on your margin for that transaction may be more than paid off in the long-term value of being able to reach that customer with more targeted marketing messages in the future.

Lapsed customers

A “lapsed” customer will mean different things to different organizations, but let’s take an example: any user who has spent money with you in the past but hasn’t visited your site in the past 30 days.

Offering these users a custom “welcome-back” experience when they return to the site could encourage them to re-engage with you more effectively. This approach could tie in very nicely with a retargeting campaign or an email blast.

High-value customers

As with “lapsed” customers, a “high-value” customer means different things in different contexts, but say, for example, that a high-value customer is one who has spent at least two times as much as the average customer in the past year. Personalizing their experience may open the door to a variety of performance improvements.

For example, you likely don’t need to incentivize a high-value customer with a discount or promotion, but consider how you can spur these customers into becoming your advocates. After a high-value user checks out, you could test a change to the “thank-you” page that invites the customer to share their purchase through their social networks or to invite a friend to browse the same product. While you may not want to ask the same thing of your typical customer, a very loyal customer may be happy to do this for you — if you only prompt them.

Merchandise affinities

If you have several product categories, you may have noticed that some customers have a strong affinity for a particular category of merchandise. For example, some customers may have made several purchases in the “polo shirts” category, but no purchases in the “jeans” category. Customers who show strong preferences for certain categories would likely have a better on-site experience if your site’s navigation reflected their tastes.

In the example of a polo shirts customer who never buys jeans, you could test your navigating to ensure that the first category listed is polo shirts. (Or you could test cross-sell techniques to get single-category customers to broaden their horizons into other categories.)

Note that this kind of targeting may require a degree of integration between Google Analytics and your organization’s CRM system, but that can be fairly easily accomplished via GA’s “Data Import” feature.

Analyzing your test results

Any testing platform will come with some built-in tools for helping you measure the results of an experiment. Google Optimize is no different: You can see how different variations are performing against goals that you have chosen in advance (as shown below), and the system lets you know when it has definitively found a winner.

But stopping there would be a mistake; you can dive much deeper using Google Analytics itself.

Imagine you’re testing out different variations of the upsell/cross-sell flow on your website. Of course, you want to know which variation works best overall — that’s the whole point of testing. But it would also be great to know if there are specific segments that respond differently to the experiences you’re creating.

For example, using Google Analytics segmentation tools, you could find out that while Variation B of an experience wasn’t the winner overall, it was the highest-performing experience for customers who have purchased at least three times. That idea — that specific segments prefer different experiences — can come up in any number of ways. Maybe customers who come in from search engines prefer one experience, while customers who visit your site directly prefer another.

Getting down to this level of detail is crucial if you want to get the most “bang for your buck” from your testing efforts. You’ll still be able to implement a winning variation, but you’ll also understand that certain segments are ripe for a more personalized experience — which, as we’ve seen with some of the examples above, you can implement using Google Optimize.

The example shown below is a common one. Here, we see that New Visitors and Returning Visitors prefer different variations. With this deeper knowledge, we can easily serve each segment the best experience.

In summary, Google Optimize hasn’t reinvented the world of A/B testing — ultimately, you’re creating variations of an existing experience and serving those variations at random to a given audience of people on your website.

Instead, the exciting part of Optimize is the ability to define and target those specific audiences in the first place. Using Google Analytics’ long-familiar segmentation tools, we can now home in on audiences that are much more specific and customized than most other tools can easily reach. Then, using those same tools, we can analyze test results to gain a much deeper understanding of what’s working best for different segments of our audience.

Like any other process, when it comes to testing, you get out what you put in. As a result, with better targeting of test audiences and deeper analysis of your test results, your tests are more likely to yield valuable insights and outcomes.

Some opinions expressed in this article may be those of a guest author and not necessarily Marketing Land. Staff authors are listed here.

About The Author

Nick Iyengar

Nick Iyengar is an Associate Director of Digital Intelligence at Cardinal Path, where he is responsible for helping his clients improve their profitability by building their analytics capabilities. He recently returned to Cardinal Path for his second tour of duty, having completed his MBA at the University of Michigan Ross School of Business last year. At Cardinal Path, Nick has led Google Analytics implementations for dozens of organizations in a wide variety of industries. Prior to joining Cardinal Path, Nick began his career in digital analytics at Google, where he managed Google’s Analytics Guru team.