Klaviyo Personalization: 7 Ways to Utilize Advanced Segments

Building targeted audiences in Klaviyo goes leagues beyond basic segmentations like “engaged subscribers” or “past purchasers.” 

By leveraging Klaviyo’s powerful segmentation features, you can create highly personalized experiences that increase conversions and drive more revenue. In this guide, our subject matter expert explains how to use Klaviyo to do advanced personalization. 

So, let’s get started with predictive analytics in Klaviyo, which is by far the most important strategic lever. 

1. Use Klaviyo predictive analytics for advanced targeting

2. Create behavior-based segments

3. Segment by engagement level

4. Personalize based on purchase behavior

5. Use dynamic segments for real-time personalization

6. Leverage email plus SMS

7. Segment based on preferences & CPD

1. Use Klaviyo predictive analytics for advanced targeting

Klaviyo’s predictive analytics can help identify high-value customers and anticipate behaviors such as:

• Predicted CLV (Customer Lifetime Value): Focus on customers with high predicted CLV and offer them VIP perks.
  
• Predicted Next Order Date: Send reminders or replenishment emails based on buying cycles.

Source: Mavlers

Take the case of predicted CLV. Now, it’s important to remember that as per the laws of statistical modeling, the value of predicted CLV is best derived from a segment-based approach, instead of an individual-based analysis. It’s far more helpful, and accurate, to interpret the CLV across a segment of customers since it paints a more statistically significant picture of customer behavior, revealing trends and patterns according to which you can reliably allocate resources. Let’s consider an example. 

If the number of predictive orders for Megan is 1.51, it posits that Megan may place one or two orders, although in reality she may make purchases wide of the mark on either side of the threshold. Putting it neatly, it is possible, but not probable, that Megan’s purchase will not conform to the predicted value. 

But, when you need to make a reliable decision, you’ll need to base it on probabilities (which are mathematically significant), not possibilities (which are no better than notions). In Megan’s case, you’ve practically got no insight, let alone a reliable one, into resource allocation. Now, if you see that Megan’s number of predicted orders is comparable to four other customers’, you can discover a scientifically sound, reliable trend (see the comparative plots below). 

Let’s say that apart from Megan, there are four other customers with a predicted number of orders of 1.49, 3.11, 0.95, and 1.12. Now you can reliably expect a total of eight orders from this segment of customers. 

With respect to tailoring emails to loyal customers, you should:

• Make the intent of the email clear in the subject line. You can state the word VIP, which is the most common way to do it. Or, you can set the customer up to receive such an email by doing a 2-series VIP access campaign, too. In either case, you’ll want to ensure that the customer doesn’t miss the email.
  
• Reveal the details of the VIP access above the fold. Using a simple GIF or bold typography can go a long way.
  
• Place the primary CTA in the hero banner, make it stand out. 

Below is an example of an email from Swoon, which offers VIP access to a loyal customer. 

Source: RGE

Swoon’s subject line reads: VIP ACCESS: LEMONADE SHORTBREAD COOKIE MIX

2. Create behavior-based segments

You can create the following behavior-based segments in Klaviyo:

• Browse Abandoners: Users who viewed a product but didn’t add it to their cart. Retarget them with dynamic product recs.
  
• Cart Abandoners: Customers who abandoned their cart several times might need a cart reminder. However, beware of repeat abandoners, you don’t want to reward the habit. Equally, don’t rush to offer discounts. Your first cart reminder should be just that, a discount-less reminder, and only sent out if you’re certain that the customer isn’t likely to return on their own. So, you want to wait before launching a cart abandonment series.
  
• Purchase behavior: Based on purchase behavior, you can create segments for one-time buyers, bargain-hunters, repeat buyers, and so on. More on this later.
  
• Loyal Customers: Those who have made multiple purchases and engage with emails frequently. Reward them with early access to sales. Just like we saw in our previous example. 

Consider browse abandoners. Once again, you’ll need to ace the subject line to begin with. Because you want the user to notice. As far as a browse abandonment email is concerned, you’ll need to be at the top of the persuasion game. While design has its role to play, you’ll benefit from a dedicated copywriter. Best not to use AI, though you can. Catch the following example from Something Navy. 

Source: RGE

The navbar is of singular importance in a browse abandonment email. 

Additionally, you can afford to keep it short since the user has already shown some interest in your products. Don’t overdo it. 

3. Segment by engagement level

You can set up the following segments to begin with:

• Highly Engaged Customers: Send exclusive offers or VIP content.
  
• At-risk Customers: Re-engage those who haven’t opened emails in 60+ days with a win-back series.
  
• Lapsed Customers: Send reactivation campaigns with personalized recommendations.

You might wonder, Why not flush out lapsed customers altogether? 

Now that’s what you may need to do eventually, but don’t start with that. As an e-commerce brand, any assumption on your part must be positive to begin with. Getting rid of “lapsed” customers from the get-go—a long-advocated best practice meant to serve ESPs, not brands—can cost you what might have been a potential source of revenue. 

As Kath Pay points out, “Marketers have many solutions available to them to win back lapsed customers without deleting their records after just a few months. Also, we know that inactive email customers might still buy from you even if they don’t open your emails.” So, you want to check your other channels too. Now here’s a cool send from Michael Kors, which is targeted at a lapsed customer(s). 

Source: Email Love

Once again, you’ll need to be at your persuasive best. The subject line and preheader assume tremendous significance here. 

4. Personalize based on purchase behavior

You’d be wise to kick off with the following segments in Klaviyo, as far as purchase behavior is concerned: 

• First-Time Buyers: Consider offering a post-purchase sequence with onboarding and product care tips.
  
• Frequent Shoppers: Upsell or cross-sell complementary products.
  
• Seasonal Shoppers: Identify customers who purchase only during sales or specific holidays and tailor promotions accordingly.

So far, so good. However, purchase behavior isn’t always a straight line. It’s multidimensional in fact. You will have to consider that customers often tend to jump stages in a buying cycle, leading you to toss out any predetermined flow. Furthermore, considering that multiple channels besides email are operating simultaneously, purchase paths can be altered dramatically. There’s also the issue of overlapping segments. A frequent shopper can also be a seasonal shopper; one can be a first-time buyer of one product and a frequent buyer of another product. 

For perspective, check out the following plot which captures effectively the differences between the linear and real-world scenarios. 

In the above, we consider the potential trajectory of a first-time buyer. 

In the linear case, the buyer goes on to make their first repeat purchase; but in the second case, the first-time buyer does NOT buy a second time from the same seller, despite reading product reviews and placing an order in the cart. The latter is the real-world scenario. What might have happened is that in the second case, the buyer shifted to a different segment altogether (from the marketer’s perspective), implying a case of overlapping segments. Maybe the customer is a seasonal buyer as well as a first-time buyer. 

In that case, a sustained post-purchase sequence won’t make sense. 

The better approach would be to study the buyer’s behavior over time, and find out if they resume activity around a peak season. 

5. Use dynamic segments for real-time personalization 

Klaviyo’s real-time segmentation updates lists dynamically, ensuring personalized messaging. For instance:

• Customers who purchased in the last 30 days receive a different message than those who haven’t purchased in six months.
  
• Users who clicked on a product in the last email but didn’t buy receive a follow-up with a special incentive.

In the previous section, we touched on overlapping segments. Klaviyo updates segments in real-time based on user actions. “While building your segment, you only have to set the conditions once. After that, the segment will continually add and remove profiles based on the criteria you’ve set,” states this help doc on dynamic segmentation. 

In fact, with Klaviyo’s Segments AI, you can simply describe the customer profile or behavior you’re looking for, and the tool will create a precise segment for you. For instance, if you want to identify customers in your database who are at a high risk of churning, you can instruct the AI as follows: “Can you generate a list of customers from my existing database who are likely to disengage or churn based on their recent interactions and engagement trends?“ And the system will list all the segments matching your request.

6. Leverage email plus SMS

Use cross-channel automation to optimize engagement. For example:

• Subscribers who engage more with email than SMS should receive different frequency cadences.
  
• Send an SMS reminder 24 hours after an abandoned cart email to reinforce the message.

SMS often gets an unfair reputation for being aggressive, largely due to overuse by pushy marketers eager for attention. However, SMS itself isn’t inherently intrusive. When you message someone who has willingly opted in, there’s no issue—but consent alone doesn’t cut it. 

Relevance matters just as much. The key is sending the right message to the right audience through thoughtful personalization. Take this SMS from Rainbow, for example. It’s crafted specifically for Ruby, a valued customer, rather than a random recipient unfamiliar with the brand.

Source: Attentive

E-commerce brands frequently share in-depth gift guides with their email subscribers during seasonal campaigns. But, not every recipient follows through with a purchase.

To capture last-minute buyers, brands can send a timely SMS reminder on the day of the occasion. Uncommon Goods appears to have used this tactic for Father’s Day—starting with an email, then reinforcing it with a brief text message and a direct purchase link.

7. Segment based on preferences & CPD

Collect preference data through surveys, quizzes, or pop-ups, and then segment customers based on:

• Favorite/most purchased product categories
  
• Size or color preferences (e.g. in the case of fashion brands)
  
• Dietary preferences (for food brands)

CPD stands for customer profile data. 

So there you go. 7 ways to create and utilize advanced segments. Make sure to test different audience segments to see which convert best. 

Keep refining segments based on engagement, purchase frequency, and campaign performance data. Now of course, you can’t directly A/B test segments; however, you can derive statistically significant results by:

• Creating custom properties (engagement levels, for example) based on which users will be routed down different flow pathways, allowing you to compare the property values across the pathways.
  
• Using profile filters on the flow trigger while comparing two email segments, so that no user receives the same email twice.
  
• Leveraging conditional splits in Klaviyo to test an element within a single flow.
  

The Most Trusted Klaviyo Pros Around!

Want to utilize such insights for your e-commerce business? Book a call with our Klaviyo-certified experts! 1000+ campaigns and automations a year. 100+ Klaviyo clients, and counting. 

Get the most out of our six-tier email marketing framework.