Segment conditions reference

You will learn

Learn more about segment conditions, including how to filter your conditions and more. When building segments in Klaviyo, conditions are the building blocks that you customize to either include or exclude certain sectors of your audience. In a nutshell, profiles will only appear in your dynamic segments when they meet the specific conditions you make when creating the segment. 

Segment conditions

A single segment can have a maximum of 100 conditions. 

When building a segment, you can include or exclude people based on the following conditions:

• What someone has done (or not done)
  You can choose from any activity tracked by Klaviyo — we call these events or metrics. To see a list of all metrics tracked in your account, head to Analytics > Metrics in your Klaviyo account.
• Properties about someone
  You can segment based on attributes associated with profiles in your account — custom properties can be assigned via an integration or you can add them to profiles manually.
• Where someone is located
  When targeting customers in the US, EU, Australia, New Zealand, or Canada, this option allows you to segment based on a certain radius of a specific zip code.

  For UK zip codes, we support filtering by outward code, not inward code or both outward and inward code (usually separated by a space). For example, if a person's full zip code is "SW1W 0NY", only the first piece ("SW1W") will work for these filters.

• If someone is or is not in the EU (GDPR)
  This option is useful if you want to exclude people in the EU from a particular segment if you haven't collected GDPR compliant consent. This segment condition does not include people located in the UK, so an additional location-based rule should be added to your segment definition in order to comply with GDPR UK legislation.
• If someone is in or not in a list
  This option is great if you want to combine 2 or more lists or create a segment of a specific list, which is a best practice when sending campaigns.
• If someone cannot or can receive marketing
  Choose whether segment members should be able to receive marketing for a particular channel (i.e., email, SMS, or push). To ensure they are subscribers for that channel, click Add filter and select Because person > subscribed.
• Predictive analytics about someone
  Predictive analytics allow you to segment based on CLV or predicted gender, which, in turn, can help you identify customers who are likely to become repeat purchasers, who are not likely to purchase again, and more. Please note that you will only see the CLV condition if you meet the criteria outlined in this article.

In the following sections, we will go through each one of these options in more detail.

What someone has/has not done

This condition allows you to segment profiles based on events in Klaviyo, which represent actions profiles have taken. If you navigate to Analytics > Metrics, you will see all of the activity Klaviyo tracks through your different integrations. These metrics are available for segmentation using the What someone has done (or not done) condition.

For example, you can find all customers that have made at least 2 purchases:

You can adjust and filter the following aspects of the metric you select:

• Frequency
  After selecting an activity, you can segment based on how often (or not often) people have done the activity. The allowed operators are:
  
  • At least once
  • Zero times (has not taken this action in the time period selected)
  • Equals (is exactly the number specified)
  • Doesn't equal (is not exactly the number specified)
  • Is at least (greater than or equals)
  • Is greater than
  • Is less than
  • Is at most (less than or equals)

• Timeframe
  After choosing a frequency, you will also need to select a timeframe over which we'll count the number of events for each person. These calculations are based on UTC. The options here are:

• • Over all time
  • In the last (within the last X days, weeks, or months)
  • Between (between 2 relative dates, i.e. between 30 - 90 days ago)
  • Before (before a specific date)
  • After (after a specific date)
  • Between dates (between two specific dates) — note that this date range will include people from the first date specified and the end-date.

If you click the filter icon, you can further hone the types of events that are included. All metrics have a unique set of properties associated with them, and you can leverage these properties to be more specific about who you capture in your segment.

For example, if you have an Ordered Product metric that captures each time someone purchases a product, you might want to filter on the Variant Name property to find people who bought a specific product variant.

The search field for values in event and profile data can display up to 1,000 unique values. If there are more than 1,000 unique values for a property within your event's metadata, it may not appear in the text field dropdown. Additionally, values in segments are case sensitive. For example, a segment of profiles with @Gmail.com email addresses would have different results than a segment of profiles with an @gmail.com email address.

In the example above, if there are more than 1,000 Variant Name properties (e.g., Chocolate, Vanilla, Mint chocolate chip), your desired value may not appear in the search field. If this is the case, simply type in the full property value (i.e., Chocolate, case-sensitive) and your segment will work as expected.

When setting a filter for an event condition in a segment, you will also be able to select an operator. The options available depend on the data type, which is automatically set to reflect the values for the event data. Be cautious when adjusting the Type field, as it may cause segments to evaluate in unintended ways if the wrong data type is selected. 

To set the data type for a condition, select the desired icon in the operator field for a condition. 

• String
  Represented by the “T” icon 

• Number
  Represented by the “#” icon. 

• Boolean
  Represented by the left and right arrows icon.

• List 
  Represented by the bulleted list icon.

When setting a filter for an event condition in a segment, you can select from the following operators based on the data type. 

Properties about someone

This condition allows you to filter based on profile properties (i.e., the information you've collected about your contacts in Klaviyo). Any data in the Custom properties section of a Klaviyo profile is available for segmentation using the Properties about someone option. 

By default, you will have access to and can use the following Klaviyo properties to build a segment:

• City
• Country
• Email 
• First Active
• First Name
• Initial Source
• Last Active
• Last Name
• Organization
• State / Region
• Unique ID
• Zip Code
• Phone number (allows for use of the Country code in filter)

As you collect additional information about your contacts (i.e. custom properties), you can use this information to build segments. For example, if you collect data on product preferences from newsletter subscribers, you can segment based on their responses.

After selecting a profile property for segmentation, you will also be able to select an operator. The options available will depend on the data type of the property, which automatically updates when you choose a property to reflect the values that property contains. Be cautious when adjusting the Type field, as it may cause segments to evaluate in unintended ways if the wrong data type is selected. 

Below are the operators for string properties:

• equals
• doesn't equal
• contains (property contains the string specified, if the property consists of text; if the property consists of a list, using the "contains" operator will search for list items that exactly match the dimension value you set)
• doesn't contain (property does not contain the string specified) — please note that this will not grab those with a value not set. For example, if you build a segment of people whose country does not contain United States, this will only pull in those with a country value set to something other than the United States, not those with the country value not set.
• is in (property is in a set of values)
• is not in (property is not in a set of values) 
• starts with (property starts with the string specified)
• doesn't start with (property does not start with the string specified)
• ends with (property ends with the string specified)
• doesn't end with (property does not end with the string specified)
• is set (property exists for the profile)
• is not set (property does not exist for the profile)

After choosing a property and an operator, the final piece is the dimension value. When you click, this value field will populate with a dropdown menu that features all available values given the property, operator, and data type selected.

If the property you select contains values that are numbers or dates, make sure to adjust the data type appropriately. Doing so will reveal type-specific operators and this will impact the values available for segmentation.

Below are the data types available:

• String: string value of plain text.
• Number: integer value.
• Date: date/time value.
• Boolean: true or false value.
• List: array of values (e.g.  ["Offer1", "Offer2"]). Note that only simple lists are segmentable; complex lists (i.e., a list within a list) are not supported in segment filters. 

Someone's proximity to a location

Use postal codes to filter a segment by the location listed in a recipient's profile. This enables you to send regionally specific information to people who live at a set distance away from a location.

If you have a brick-and-mortar store or an in-person event, this allows you to target your customers and brand enthusiasts who may be interested in attending.

Even if you don't have a physical store, you can use this feature to send information about regionally specific products (i.e. gloves in cold places during the winter) or send promotions to a certain area. Additionally, this allows you to use a signup form to create a popup that will display to a particular region, allowing you to personalize your recipient's experience on your website.

If someone is in or not in the EU (GDPR)

This condition can be used to identify people in the EU. This is useful if you would like to exclude these contacts from a particular segment to comply with GDPR.

If someone is in or not in a list

This condition can be used to segment if someone is a member of (or not a member of) a specific list. 

Use this segment if you want to capture a cross-section of that list. For example, the following segment will identify everyone in your newsletter list from the United States:

When segmenting based on whether or not someone is in a list, you can filter based on when that person was added to the list:

• in the last (within the last X days, weeks or months)
• between (two relative dates)
• before (a specific date)
• after (a specific date)
• between dates (two specific dates) — note that this date range will include people from the first date specified, but not the end-date. Because the last date isn't actually included in your segment, add one day past the end-date to your time range in order to include your desired end-date.

If someone is or is not suppressed

In October 2023, the condition If someone is or is not suppressed for email was replaced by the condition If someone cannot or can receive marketing. Segments were automatically updated to use the new conditions, and this change had no impact on segment membership. Learn more about the transition to the new segment criteria.

If someone is or is not consented to receive SMS

In October 2023, the condition If someone is or is not consented to receive SMS was replaced by the condition If someone cannot or can receive marketing. Segments were automatically updated to use the new conditions, and this change had no impact on segment membership.

If someone cannot or can receive marketing

Segments pull from all the profiles in your account, regardless of the marketing channels they are or are not subscribed to. When you create a segment, it likely contains some profiles that are suppressed for a certain channel (e.g., email), meaning you may not contact them via that channel.

To get an accurate picture of how many segment members you can contact, use an AND connector to add this condition: If someone cannot or can receive marketing > can receive > email marketing (or SMS marketing, or mobile push marketing).

If you’d like your segment to only contain profiles that are consented (i.e., opted in) to a marketing channel, add the condition above. Then, click Add filter and select because person > subscribed.

Predictive analytics about someone

This can be used if you would like to segment based on customer lifetime value (CLV) or on their predicted gender. When using this the Predictive analytics about someone condition, you'll have the following options:

• Predicted gender
  Klaviyo’s gender prediction algorithm uses a customer’s first name along with census data to make a gender prediction of either likely male, likely female, or uncertain.
• Historic CLV
  This is the sum of all the previous purchases an individual has made.
• Predicted CLV
  This is a prediction of how many purchases a customer will make in the next year. It's not an exact prediction, so in some cases, you may see an impossible number of predicted orders. For example, you may see 1.43 as the number of predicted orders for a particular customer. This means we expect them to make one or two orders, but there’s a chance they'll make more or fewer. These expectations start to make sense when you group multiple customers together because you can predict the total number of orders or spend for the group. If we have five customers with a predicted number of purchases of 1.43, 0.25, 3.12, 0.78, and 2.97, we can expect approximately 9 purchases across this group.
• Total CLV
  This is the sum of the Historic CLV and Predicted CLV of a customer.
• Predicted number of orders
  This is the predicted number of orders a customer will place over the next year.
• Average time between purchases
  This is an average of the number of days that have elapsed between each of a customer's purchases.

When building these segments, you can restrict your desired value to:

• is at least
• is at most
• specified based on gender or uncertain

Filtering a segment condition

When building segment conditions, you have the opportunity to use filters to refine your segment. You can do this by choosing a segment condition, and then selecting the filter icon. 

You will then see an input where you can select a property to filter on. After choosing your property, select a value for that property. A dropdown menu will also populate here with available values. The options in the dropdown are only those that have synced to your Klaviyo account.

You can also include multiple filters on a single event condition. For example, you can include profiles that placed an order with a certain number of items, and one of those items is from a particular brand. To include multiple filters for an event condition, select the filter icon again. 

You can have a maximum of 10 filters for each event condition. Once you meet the maximum number of filters, the filter icon will no longer be present for the event condition. 

While Klaviyo may sync many details about a given metric, not all synced properties are available for segmentation. For data management purposes, only the primary details of an event are synced as top-level properties, and you can segment only these top-level properties.

If you view the raw data Klaviyo syncs for an event, you will see key data points for the event. For a Placed Order event, for example, you may see the following top-level properties:

• Value
• Collections / Categories
• Item Count
• Items
• Source Name

You will see an array labeled "Extra" or "Details." While the data within an "Extra" or "Details" array is available to insert within an email template, you cannot segment based on properties nested within this array.

Custom integrations & segmentation

If you're building your own custom integration and deciding how to structure the data that you're sending to Klaviyo, make sure that all properties you want to segment on are sent as top-level properties (and are not nested).

For example, if you create a Placed Order metric where an event will track whenever someone completes a purchase on your site, you may want to filter events based on what item(s) someone purchased.

In order to do this, there has to be a top-level "ItemNames" array that includes all purchased product names — this would be in addition to a nested array with details for each item. Below, you can see the ItemNames array in the payload for this Placed Order event in addition to the Items array that contains detailed information about each product:

{
  "token" : "API_KEY",
  "event" : "Placed Order",
  "customer_properties" : {
    "$email" : "john.smith@gmail.com",
    "$first_name" : "John",
    "$last_name" : "Smith"
  },
  "$event_id" : "1234",
  "$value" : 29.98,
  "Categories" : ["Fiction", "Classics", "Children"],
  "ItemNames" : ["Winnie the Pooh", "A Tale of Two Cities"],
  "Brands" : ["Kids Books", "Harcourt Classics"],
  "Items" : [
    {
      "SKU" : "WINNIEPOOH",
      "Name" : "Winnie the Pooh",
      "Quantity" : 1,
      "ItemPrice" : 9.99,
      "RowTotal" : 9.99,
      "ProductURL" : "http://www.example.com/path/to/product",
      "ImageURL" : "http://www.example.com/path/to/product/image.png",
      "Categories" : ["Fiction", "Children"],
      "Brand" : "Kids Books"
      },
    {
      "SKU" : "TALEOFTWO",
      "Name" : "A Tale of Two Cities",
      "Quantity" : 1,
      "ItemPrice" : 19.99,
      "RowTotal" : 19.99,
      "ProductURL" : "http://www.example.com/path/to/product2",
      "ImageURL" : "http://www.example.com/path/to/product/image2.png",
      "Categories" : ["Fiction", "Classics"],
      "Brand" : "Harcourt Classics"
    }
  ],
  "time" : 1387302423
}

For more information on Custom Integrations, check out our article on manually importing historical event data.

Additional resources

• AND vs. OR guide
• Getting started with segments