This introduction to segmentation and testing provides key conceptual explanations and introduces the member to various approaches that can be found in Google Analytics to resolve data needs.
- [Narrator] The practice of segmentation in marketing literally means to divide a large homogeneous market into smaller segments that share similar characteristics. With conventional marketing, the basis for segmentation are often limited to below eight, but with digital marketing, BigCommerce and Big Data, though literally hundreds of basis for segmentation. For example, device category, install location, geo location, customer language, transactional value, and so forth.
As you can see here, Google Analytics makes a good attempt at providing us with various pre-find options that can be applied to submit our data. But at times, we can use other functions and views to resolve our data needs. For example, segments can also be obtained by setting up filters in the Admin area of Google Analytics. The whole idea of segmentation is that the size of each potential segment can be measured.
Then based on the potential value and the size of the segment, we can adjust our marketing mix for each segment. Let's imagine our analytic data tells us that we have more high value customers in California than in Texas. Then we can increase our marketing budget in California and even create a more localized landing page for our California clients. Now let's imagine that analytics data tells us that middle-aged mobile users in India found us via Bing and they are more interested in our product than middle-aged desktop users who found us via other search engines in the U.S.
In this case, what we might do is to create specific ads for mobile users in India with an increased budget for that particular segment of users on a specific channel. We can also specify a hypothesis, and we can test it. For example, we may believe that our segment of 2,000 mobile visitors per day will engage better with our content if we use smaller images and then test it to see whether the bounce rate for this particular segment have decreased after we made the change.
Or we might also believe that our segment of 50,000 daily visitors in India will respond better to a different value proposition, which will reflect in our online copyrighting. So, we can make this change, and we can test if it does lead to a better conversion. The ability to analyze different segments of users and to test different marketing propositions can help to redefine reciprocal value between the business and its customers.
So here is something for you to think about. Can you think of potential segments of users in your business that'll benefit from a different marketing mix? And then, how can analytics be used to reveal more about these potential segments?
- The difference between planned and retroactive segmentation
- Setting up segments for testing in Google Analytics
- Using demographic segmentation to plan future advertising
- Testing the popularity of website content with behavioral segmentation
- How analytics reveal social segments
- Testing segment conversion rate and performance
- Achieving segmentation with conversion segments within multichannel funnels
- The differences between multivariate and A/B testing
- Creating a digital marketing report