What is Google Optimize?
Google Optimize is an analytics tool built into the premium Google Analytics service (Optimize 360). On September 28, 2016, we received a notice from Google stating that they will be rolling out a FREE version of Optimize and that we can start beta testing it today!
This is exciting news, especially since we know a ton of clients that can benefit greatly from this service but could never justify the cost until now.
Google Optimize is designed to enable publishers to maximize their engagement with customers. With this analytics tool, you can do the following in a fraction of the time that it would otherwise take:
- A/B Test
- Cater content based on customer metrics
- Apply Bayesian modeling to make smarter business decisions
- Target customer’s more efficiently
- Acquire more data, more knowledge and more insights, faster
Testimonials touting the success of this product are easy to come by, even from large well-known companies like the Motley Fool and AccuWeather. It may be no surprise that these companies work in very data-driven industries (stock market and weather).
How You Can Use Google Optimize
There is a lot you can do with this tool along with the suite of other tools that come with Google Analytics, but a few specific strategies are very important to us (Strunk) and our clients, namely…
A/B Split Testing
What I love about A/B testing is that it’s simple and complex at the same time. You can use this technique to make some very basic choices and have immediate payoffs, or you can drill down to an almost inane level of detail and try to perfect every aspect of every process. Conversion funneling has never been so effective before and it’s easy than ever to make smart decisions that even Google Would Do.
With Google’s easy interface, you can set up totally different versions of the homepage or make minor subtle variances of a landing page. Compare different content, Call-to-actions, messages, displays, interactions and more with a few simple steps rather than trying to re-invent the wheel.
We apply this strategy all the time. So does Google themselves, along with Amazon and every other major tech giant in the industry, you should too.
Advanced Statistical Modeling
This is a very useful tool when you want to appear fancy in front of your clients, like we do. Using Bayesian statistical methods, we can display some pretty interest data in a real-world visualization to simplify the results of experiments that you run. And while it does give us a chance to look smart, it IS smart.
Data is not very useful if it’s not accurate or modeled properly. The Google Optimizer tool can analyze the results of tests and experiments for you and even provide what sort of actionable items can be done to capitalize on this information.
While it’s fun to use obscure mathematical jargon to impress clients, the concept behind Bayesian statistics is simple. Beliefs are expressed as probabilities which change according to new data.