This article is part of our series on customer experience where we look at how to connect data, intelligence, and experiences. Read the previous article, Great Customer Experiences Rely on Robust Identity Management.
Your ability to deliver personalized customer experiences across multiple channels is increasingly being tested. There is a good reason for this – cross-channel communication is more effective, harder to do, and demanded by your customers. It's a fact: Targeted campaigns deliver much higher conversion rates than batch-and-blast efforts.
Unfortunately, the foundations upon which these experiences are built – data, segmentation, and the insights they generate – are too often poorly designed and even more poorly integrated.
For example, data is often trapped in silos, preventing any linkages and the insights this can produce. The ultimate result is incomplete or poor segmentation and, in turn, disconnected customer experiences.
If either the data or the intelligence applied to it are lacking, the result will be substandard. That is because disconnected data combined with disconnected intelligence produces disconnected experiences.
Breaking down organization data silos is the first step, but it’s not an easy one. This is particularly challenging for organizations dealing with complex or outdated legacy systems. However, success on this front creates a valuable resource for business and one that can underpin targeted marketing campaigns.
Of course, data alone lacks utility. The question for marketers is how to pool the data, and then create useful insights, which can be turned into appropriate messaging for customers and prospects.
Segmentation may sound like marketing 101, but it is a process that still trips up some marketers. Effective segmentation is foundational to any marketing or advertising strategy and spans a range of tactics and meanings.
At a simple level, it is essentially labeling groups of people based on behavior, demographics, marketing tactics, and personas.
The temptation is to segment once, establish your customer labels, and then move on to insights. But effective segmentation is a continuous, iterative process that enriches your data over time. That also means segmentation must remain attached to the data and the technology stack.
The trap many marketers fall into is slicing their customer base into labels without connecting the technology stack. While the labels may be based on sound data insights, failing to tie them back to the data creates problems as it is unclear what characteristics make up the labels.
This is known as aspirational segmentation, and it can create problems when it comes time to execute. For instance, you might find that when you run a campaign to a certain segment you can't actually find that group of people because they are not connected to the technology stack.
Let’s say your target group is ‘fun-seekers.’ Marketers need to know exactly which attributes or data points qualify a prospect as a fun seeker. Do they frequent music festival sites? Listen to a particular genre of music? Or perhaps take advantage of travel promotions? However marketers wish to define it, knowing this allows for more effective targeting, but also fast tracks increasingly automated messaging.
If the funseeker segment isn’t attached to data, it's almost impossible to target, and without data it is also subjective.
And getting it right will give you foundational knowledge of your customers, which can be leveraged in effective data-driven marketing.
If segmentation is data-driven and data attached, then it can be executed upon and it's part of the technology stack. So not only can you campaign, but you can also run reports. Then, you can test and tweak it, and ultimately optimize the campaign.
Segmentation means lots of things. It can be as simple as selecting groups of males or females. More data complex segmentation might involve merging multiple data sources and using data science and machine learning to build a customized segmentation system.
The important thing is segmentation remains attached to data and becomes executable.
Marketers should aim to segment all their data, but it doesn’t have to happen all at once. Start small, segment the data you have access to, and see what works. Expand when you see the results. The most important thing is to get started!
Download our free Busting 5 Common Myths of Marketing Automation to learn how the right segmentation is the start to proper predictive analysis, account-based marketing, lead nurturing, and attribution modeling.