AI segmentation: 3 best practices for saving time and powering personalized marketing

Profile photo of author Jessica Schanzer
AI at work
August 5, 2024
In dark lavender font on a lighter lavender background, capital letters read, "AI segmentation."

You’ve heard it before: consumers are raising their standards about the brands they do business with.

And personalization is a big part of that equation.

According to McKinsey & Company, 72% of shoppers expect the businesses they buy from to recognize them as individuals and know their interests, while 78% say they’re more likely to make repeat purchases from companies that provide a personalized experience.

When you segment your customer base into distinct groups based on their behaviors, demographics, and even predictive analytics, you’re better equipped to tailor your marketing messaging to only the most relevant content that resonates with their unique needs and preferences.

But one of the biggest challenges of smart segmentation is the time-consuming process of collecting and preparing your customer data.

This is where AI segmentation comes in.

The typical ecommerce business has audience data in many different places, from CRM systems to website analytics platforms. After collecting all this data, you still have to extract the data points relevant to your segmentation criteria. It’s a lot of work, especially if you have a large audience.

AI can analyze huge volumes of data and identify patterns and correlations much more quickly than humans can. This minimizes a key challenge of customer segmentation using AI and empowers you to create granular segments based on specific customer behaviors, purchase history, and interactions across different channels.

Some AI segmentation tools even have predictive capabilities and can forecast future trends or customer behaviors based on historical patterns. This enables you to anticipate customer needs and tailor your offerings and messages to capitalize on emerging opportunities, keeping you a step ahead of competitors.

With Klaviyo Segments AI, simply describe the groups of people you want to target, and Segments AI creates the segment for you.

Here are 3 best practices for using AI segmentation to tackle different steps in the process.

In a data privacy-forward world, it’s crucial for marketers to adhere to email consent laws and anti-spam regulations.

That’s why, in Klaviyo, if you’re building a segment you plan to use for any of your marketing channels (such as email or SMS), you need to include a consent statement in your prompt.

For example: “I want people who consented to receive email and SMS marketing and who have viewed a product and not made a purchase in the last 60 days.”

Note: Klaviyo Segments AI doesn’t automatically add this consent condition because you might be building a segment where marketing consent isn’t relevant (like a segment for a transactional email).

2. Use AI as the starting point for new segments

Creating audience segments isn’t necessarily difficult, but it can be time-consuming. To save time, use Klaviyo Segments AI to instantly turn a few prompts into a new segment.

Segments AI is very customizable—it lets you get as narrow and specific as you’d like when it comes to grouping subscribers. For example, your prompt might be “Everyone who lives within 150 miles of Austin, Texas.”

Using natural language prompts like this one, you can prompt Klaviyo to create an audience segment based on a variety of criteria, from location to churn risk.

Image shows a screenshot of the Segments AI builder in the back end of Klaviyo, titled, “Define with AI.” The user inputs the name of their desired segment, then writes in a description of the segment (in this case, “Everyone who lives within 150 miles of Austin, Texas”). The options then include “Preview segment” or “Try an example.”

Note: Segments AI is only available if you’re building a new segment. If you want to edit an existing segment, use the dropdowns in the segment builder. Using Segments AI mid-build will overwrite the criteria you have already entered into the segment builder. Always use Segments AI as a starting point, then add or modify conditions as necessary afterward.

3. Check AI’s work

AI is a great tool that can give you a major head start. But it’s not always foolproof, and you still need to get yourself past the finish line.

Whether you’re using AI to generate content or analyze data, it’s important to have a human being check AI outputs before launching related marketing efforts.

Segments AI is designed to help you hone in on your desired criteria faster, but you get the final say over your segment criteria. Always review AI-generated segments in case you need to add or modify any conditions in the segment builder.

Start using Klaviyo Segments AI today

AI customer segmentation is a powerful tool that can save time, increase efficiency, and improve the effectiveness of your marketing efforts.

By using the speed and predictive capabilities of Klaviyo Segments AI, you can quickly create highly targeted segments and send them the kind of content that drives engagement, conversions, and customer loyalty.

Related content

Segment smarter with Klaviyo.
Get started
Jessica Schanzer
Jessica Schanzer
Product Marketing Lead
Jessica is a product marketing lead at Klaviyo, focused on helping customers succeed with AI. She’s located in Austin, Texas and has spent the last 5 years in ecommerce. In her free time, Jessica can be found rock climbing, training for her next medieval LARP event, or spoiling her lizard Charlie.