Kibo: We agree that the crawl, walk, run approach is a smart one.
The start of the personalization journey is about learning and seeing incremental revenue gains, the end goal is to become a customer experience leader. The tools we use at the beginning are testing and analytics, next up (alongside testing and analytics) we have rules based segmentation but, segmentation only gets us so far. Segmenting your biggest most lucrative segments is very worthwhile (VIP customers is a great place to start) but, as you then look to group smaller, less valuable audiences into segments the effort increases and the rewards decrease. Here’s where (alongside the previous tools), we need to lean on the power of machine-learning to deliver personalization at scale.
Visitor intent changes and groups of customers are rarely rigid. Adding to that fluidity, we have a huge array of data points (location, demographic etc), and digital elements (layout, messaging, recommendations etc) to personalize, segmentation can be overwhelming. A solution with 1-to-1 capabilities is invaluable as it enables you to ingest all of the data on a visitor in real-time and deliver an experience most likely to result in the metric you’re trying to improve. Segmentation can’t match that level of performance.
PeakActivity: One of the biggest misconceptions around personalization is that it requires each individual user to have a truly unique site experience. Not every business has the data to create Amazon-level personalization, and it’s not always necessary either. We suggest a crawl, walk, run approach. There’s actually a lot you can do in terms of presenting more relevant messages, content, and products simply based on data collected from a visitor’s IP address. For example, you can customize a homepage banner to say “welcome back” if you know they’ve visited before, or you can highlight cold-weather products only in certain geo-locations. For more advanced clients, segmentation is a great way to create a more personalized site experience. Purchase behavior, browsing behavior, brand engagement, site visit frequency, and order frequency are great data points to use to segment customers and influence personalization.