Uncovering patterns and trends in behavior is a key step for marketers who are analyzing, optimizing or personalizing web sites. This ongoing effort can be achieved by creating segments in their analytics tool. Unfortunately, the standard, ready-made segments that many web analytics tools offer are too limited. In this article, I will outline practical segmentation ideas that you can use to find actionable insights to fuel your personalization or optimization plan.
Segments allow us to narrow down the volume of data to extract the essence hidden within. I’ve recently written about the importance of segments as a tool that allows us to tell the real and interesting stories behind the “what” and “who” of our data. Nevertheless, it is a tedious, time-consuming task. Locating distinct and powerful user archetypes, not to mention incorporating those segments into your overall marketing strategy, requires a deep understanding of who your audience is.
One suggested technique that can be used to make the segmentation process more insightful and effective is to form highly relevant persona-inspired segments. Amy Schade, Director at Nielsen Norman Group, identified five distinct types of e-Commerce shoppers:
Advertisement
-
Product-focused shoppers: Highly focused, goal-driven shoppers who know exactly what they’re looking for. These can be either new or returning customers. The key metric would be time-to-purchase at the session level. Since they know exactly what they want, they tend to purchase quite quickly. To pinpointthese shoppers, create a segment for sessions with a really short time-to-purchase.
-
Browsers: Leisurely shoppers who are looking to be inspired or are simply killing time. Browsing customers tend to spend a relatively great deal of time on the site. To identify these browsing sessions, look for unconverted segments with long session durations. These visitors may convey a huge opportunity for you to influence their decisions and turn them from browsers to buyers.
-
Researchers: Highly motivated, goal-driven shoppers who are looking to learn and become educated about a specific product. These customers tend to spend a lot of time exploring specific product pages, learning about features, prices, delivery terms, etc. To detect the researchers, look for sessions with a relatively long time spent on product pages. If you’re offering any in-depth product materials, such as explanatory product videos, how-to articles, review articles, and so on — these shoppers may be spending time following these pages, as well.
-
Bargain hunters: Customers who are on the lookout for the best available deals. To locate these customers, look for common bargain-hunting behaviors, such as repetitive visits to products on sale, customers with coupon redemptions, customers who are sorting category pages by price, etc.
-
One-time shoppers: Customers who are not entirely familiar with your site. These could include first time visitors with characteristics similar to bargain hunters, researchers, product-focused shoppers and even browsers. If you’re offering checkout without registration, these customers are more likely to buy from your site. To uncover these users, segment single-session visits. You can explore this segment deeper by comparing unconverted versus converted sessions.
Instead of looking at endless ways of segmenting visitors, focus your efforts on these five personas and think about the distinct characteristics that represent each group of visitors. Best-selling authors Jeffrey & Bryan Eisenberg suggest to take it even further by encouraging Marketers to combine personas and storytelling to implement a more agile process. Story-based personas can convey narratives of how each type of group is engaging with your brand and impacting your business.
Of course, if you’d like to extend your efforts, there are other ways of segmenting visitors. Here are a few examples of valuable metrics that you can use to differentiate between different groups:
-
Revenue impact: Discover high-value customers by profiling your highest yielding customers. For example, focus on basket size as your metric, capture the top percentile of your top spenders by segmenting visits with a relatively high revenue per session, visits with orders of over $XXX in revenue. Analyze the behavior of this key segment and act accordingly to optimize and personalize experiences, encouraging these customers to spend even more. A deep understanding of the characteristics of these personas may help you motivate other customers to become top spenders as well.
-
Engagement or action taken: Measure different engagement levels with your brand and segment visitors by looking at specific groups, such as video watchers, visits with social media interactions, type of product engagement and so forth. Once you have this data, you can start optimizing your site based on it. So, for example, if you find that video enthusiasts are a highly profitable segment, you can delve deeper into the data and segment additional sub-groups, based on the number of videos watched. By doing that, you may find that visitors with three or more video plays yield the highest revenue, compared to visitors with just one or two video plays.
-
Type of content consumed: Categorize content based on brands, categories or type of page. For example, segment customers who have shown interest in specific brands, categories (or sub-categories), products, and even product attributes, such as specific colors or price ranges.
-
Visitors with past conversions versus no conversions: This is a fairly classic, but important, segment to look at. Instead of just looking at all of your segmented visitors, segment converters and non-converters based on traffic sources, such as paid versus organic search, display advertising, specific referrals and so on. Make the most out of these reports by locating traffic sources with a high revenue impact.
-
Customer intent: Categorize different type of groups by looking at intent cues, starting from research, to purchase. For example, segment researchers by locating unconverted visitors who may be at the early stages of the funnel, learning and collecting information about your products, promotions or prices.
To Conclude
I hope that with these segmentation ideas, you’ll be able to extract the maximum value from your data. Now over to you: Which segments are your favorites? Which ones turn out to be the most effective for you? Got any segmentation tips you care to share? Feel free to comment below!
Yaniv Navot is the Director of Online Marketing at Dynamic Yield, an SaaS-based solutions provider for real-time personalization and optimization. Navot is an experienced digital marketing expert, with vast experience in SEM, web analytics and conversion optimization.