Ecommerce Personalization Blog
Ecommerce tips, strategies, and news – all without ever having
Behavioral segmentation creates personalized customer journeys and increases profits.
This post shares advanced behavioral segmentation tactics and strategies for eCommerce brands. Today, personalization software gives you the ability to observe and adapt to customer behavior.
Adopting these strategies will not only create a better customer experience, but also make your store more profitable. if you'd like to skip straight to the behavioral segmentation examples, click here.
What is behavioral segmentation and how can we use it?
If you are already familiar with behavioral segmentation, feel free to skip to the next section.
If you're not, it is helpful to define what is behavioral segmentation.
Segmentation is the process of "dividing a market into groups of customers with similar needs, and developing marketing programs that meet those needs".
Behavioral segmentation focuses on dividing customers by their past behavior.
Because actions demonstrate more about customers' intent and preferences than mere demographic information, behavioral segmentation is more predictive, and therefore effective.
“...behavioral segmentation is more predictive, and therefore effective."
Data enables behavioral segmentation.
While segmentation as a marketing strategy has long existed, it hasn't been possible to track customer behavior accurately and at scale until recently.
Personalization software like Barilliance automates data gathering for you. Now, you can define any number of segments based on behavior, and even define specific pricing strategies for each segment.
The rest of this post will focus on successful behavior segmenting strategies.
How to create behavioral segments
An effective segmentation identifies a distinct group of customers.
We created a variety of behavioral segmentation formulas that ensure you are creating effective segmentations.
1. The base behavioral segmentation formula: (behavior) = segment
The first behavioral segment is the "base" formula.
It is the simplest type of behavioral segment you can create. It is defined as the behavior itself.
How Barilliance automates behavior segmentation
Barilliance automatically tracks user behavior on your site. To create a base behavior segmentation, simply specify which behavior you would like to create the segment on.
Options fall into three general categories.
Engagement behaviors: This includes behaviors such as distinguishing between first time visitors to returning ones, adding specific or any number of items to their cart, or viewing a product.
On-site behaviors: Engagement behaviors are a subsection of more general on-site behaviors. Other on-site behaviors include visiting specific URLs, how long they spend on your site, and specific queries they make on your site.
Traffic source: The traffic source category describes previous behaviors of your visitors before they reach the site. They are excellent at providing context. Examples include what types of offer are they responding to, what mediums do they spend time on, and what if any campaigns are tied to the session.
Visitor Insights: These are advanced cross sections, often including multiple behaviors and eCommerce KPI thresholds. We will describe these more in detail below.
Time on site > 10:00
Users who actively shopped around the site
# purchased items > 0
Users who previously purchased
2. Behavioral Segmentation - Adding a threshold: (behavior) + (threshold) = segment
We can greatly increase how precise our segmentation is by adding what I call a threshold.
A threshold is a tipping point in a specific metric. It can either be a lower limit or an upper limit, and the metric can be any metric of choosing.
An example of a threshold is a minimum of $50 (lower limit) AOV (metric).
Combining a behavior with a threshold gives you greater control on who is included in the customer segment.
Time on site > 10:00
#viewed items > 5 in x category
Active shoppers who expressed interest x category
# purchased items > 0
AOV > $50
Users who previously purchased with an AOV > $50
One common challenge with implementing thresholds is making sure your data is unified.
In fact, data unification is one of the primary criteria we outline in selecting a personalization software parter.
In this article, we describe the challenge of data unification more deeply, but for now, realize that you want to combine purchase history from all sources to create accurate segments.
3. Adding a where clause: (behavior) + (threshold), (where) = segment
The last formula we will explore today adds a "where" clause.
Recency matters. It is one of the main factors in RFM analysis, and can be incorporated into behavioral segmentations easily with the right technology.
An often overlooked segment is one that immediately purchased or experienced your site.
Another interesting implementation is to target holidays like Black Friday or seasonal shoppers. Below we provide a few examples of segments that implement a behavior, threshold, and where clause.
Time on site > 10:00
Purchase Rate > 0
Session is within the month
Users who actively previously purchased and shopped around the site in the last month
Bought items > 0
AOV > $50
On black Friday
Users who bought on Black Friday with > $50 AOV
How Barilliance implements where clauses for behavioral segmentation
Implementing where clauses is built into Barilliance.
Many behaviors have options to put in a specific timeframe. Below, you can see the option on how to create a behavioral segment based on purchase history. You can set time frames from their first order or from their last order.
With this flexibility, you can easily create segments for first time customers who naturally roll out of the segment after a certain number of days have passed.
Advanced behavioral segmentation examples
We've collected examples of behavioral segmentation from some of the top eCommerce companies.
1. Above the fold content personalization ft. Thrive Market
The most productive space to personalize is above the fold content.
Thrive Market segments customers based on previously purchased and viewed items. Here, we see that the above the fold recipe recommendations all follow the same diet category the customer is a part of (Paleo). This data is used to keep offers relevant and increase conversions.
2. Dynamic product recommendations based on in-session actions ft. Third Love
In-session actions provide the best signal for what customers are interested in now. Using this contextual data, brands can create better offers.
Below, Third Love uses dynamic product recommendations on the product page based on the currently viewed product.
3. Selectively use discount campaigns to drive non-converting leads ft. Stitch Fix
If leads do not convert on your initial offers, you should resegment them to a "non-converting" behavioral segment.
These leads can be offered more aggressive discount campaigns or be given other incentives to complete their first purchase. Stitch Fix provides a great example.
After a customer completes their initial fit assessment, they are gives a series of offers to complete their first purchase. However, if the lead hasn't bought, they are placed in a non-converting behavioral segment and a new activation campaign is triggered - this time offering a $35 credit to be used.
Behavioral segmentation is a powerful tool to create relevant marketing campaigns and content personalization throughout customer journeys.
By understanding what specific segments are interested in, you have the ability to build offers that resonate. I highly recommend learning about RFM analysis to have a core understanding of how to create profitable segments.