Customer data can be a valuable asset for brands looking to create personalized and engaging experiences. The hardest part is knowing how to properly utilize it. The use of customer data in marketing is becoming increasingly sophisticated, with new technologies and techniques emerging every day.
In this post, we’ll explore the many ways that customer data can be used in marketing and very specific ways you can get started today.
Omnichannel Integration
Omnichannel integration is the practice of sharing and reacting to data between marketing providers used to managed our channels. It’s the cornerstone of any good customer data strategy and can limit marketing opportunities when not in place.
The main benefit is to use customer data to react to different parts of your marking ecosystem. Retargeting is a more traditional example of this. I also see it used a lot in apps that leverage push, email, text and sometimes even paid media to get users back in. At a more advanced level, you could use customer data to develop propensity models to help prioritize segments at different stages in the journey.
The easiest way to get started here is to make sure all of your webforms have connectivity to your CRM. Aside from the contact form data, you may also be able to send additional user event data to your CRM. The event information may already be supported by your contact form provider or CMS, but when it’s not, it’s a relatively simple integration for a developer.
Customer Journey Mapping
Customer journey mapping is a useful way to understand and improve the customer experience. It involves breaking down your audience into unique segments representing their behavior and analyzing all of their touchpoints, to identify areas of friction and opportunity.
For example, imagine an outdoor sports retailer wants to analyze their ecommerce funnel for areas of hesitation or drop-off. One segment they may analyze could be out of state purchasers on mobile as a representation of users that might be looking for gear while visiting on a trip. You would plot out all touchpoints in this specific segments journey, including all paid/owned/earned channels and review the data at those individual touchpoints.
The hard part is usually consolidating the data in a usable yet secure and compliant manner. The good news is you can do a lot with anonymized data in the latest iteration of Google Analytics.
Segmentation
Speaking of segments, segmentation strategies are finally leaving their silos and have moved to the warehouse. Customer data platforms of various shapes and forms have become pretty accessible from a usability and financial standpoint.
A common pain point has often been how to effectively recreate audiences across platforms, for example CRM and web analytics contain very different data sets. Luckily, web analytics platforms have become more sophisticated and most now allow us to send additional information about our users to help us more accurately represent them analysis. There are also more integrated off the shelf solutions that address the issue by centralizing all of your marketing initiatives into one platform. It’s often just enough however to send the proper signals to the platforms that specialize in their specific functions.
Maybe this is more of a personal preference, but marketing service providers have become very interoperable, with integrations out of the box for popular data warehouses and other service providers. For this reason, I prefer the “roll your own” approach where you choose the best platforms for your business and let those systems talk to each other.
However you go about it, there are plenty of ways to leverage your customer data to build more defined segments that can be analyzed and activated across your entire marketing ecosystem.
At the very least, here is how to do it in Google Analytics.
Personalization
If I had a nickel …
But seriously, it seems like everyone talks about it, but aside from the obvious, we can do a lot better. Obvious personalization is mostly the simple stuff that these days seems, well, obvious. Like personalizing transactional messaging with names, location and interests or doing IP lookup to personalize a web page. Basically, it’s the kinda creepy stuff.
This is mostly because of the general complexity involved in creating personalized experiences that truly affect behavior. Machine learning and AI solutions have been working to address the problem to make these personalized experiences more accessible.
Recommendation engines seem to be at the forefront and in most cases, are the easiest to turn on. Most popular ecommerce platforms will offer an AI assisted product recommendation engine or at least the ability to integrate with a third party. An incredibly underserved audience could dramatically benefit from a properly implemented content recommendation engine, the Searcher. These people deserve a break. They’ve had to deal with crappy search experiences for too long.
Testing & Optimization
If I could, I’d test everything I ever put out into the world. Just because you can’t delete something from the internet, doesn’t mean that thing can’t adapt and change over time. No matter how hard we try, our conscious and subconscious biases are always going to be at play. I talk about it a bit more in my post about thinking big in technology. My point is we need to get out of our own ways and start optimizing experiences over time.
This is a great place to address areas of friction or opportunity that were identified as part of a user journey mapping exercise. User data can be used to identify touchpoints with the largest opportunity for improvement. You can then implement a test at that touchpoint to improve the user experience for your entire audience or one of its segments.
Customer Feedback
The easiest way to get started here is to gather feedback from your customers. In most cases, a social presence or an easy to find contact form is enough. It can also be helpful to add a dropdown to the form to help users categorize their feedback so that you can use that data to build segments within your analytics platform to contextualize it.
The issue becomes reacting to customer feedback at scale. Additional customer experience metrics attempt a more quantitative approach by measuring loyalty, satisfaction, effort and lifetime value. These can be valuable data points when reviewing various touchpoints and trying to understand their real world impact on specific segments of your audience.
On top of that, natural language processing algorithms coupled with machine learning can help speed up the more manual process of reviewing user feedback. This is mostly helpful for categorizing feedback that could tie back to specific segments, touchpoints or themes that you need more information about.