Edit

Use natural language to create segments

  • Article

Note

You will need your administrator to turn on the feature switch for the Natural Language queries for Segmentation functionality under Segmentation before you are able to access it.

Important

The the natural language feature for segments currently only supports English language queries.

If your administrator has enabled the natural language feature, you can use Natural Language Query (NLQ) to build segments in Dynamics 365 Customer Insights - Journeys. Natural language input enables you to use common words and phrases to describe the people you want to include in dynamic and static segments without learning all the specifics of the Common Data Model (CDM) and best practices for building queries.

Using the instructions in Working with segments, type and submit a statement, and the service will fill the query fields using information collected from your contacts, accounts, and leads.

Sample segments

You can create a segment that will evaluate your Dynamics 365 organizational database for contacts that meet certain criteria. For example, contacts living in a certain area who have an email address.

You might name the new segment “Contacts who live in Chicago with email and an upcoming birthday” and then define the segment as follows:

  • Contacts who live in Chicago with an email address and with a birthday in the next month

Example: Contacts who live in Chicago with an email address and with a birthday in the next month

Building the segment

  • Remember to enter the phrases using normal language. You could enter, "Contacts who live in Chicago with an email address and with a birthday in the next month."
  • Give the segment a descriptive name that will help you find it later.

Reviewing the results

  • After going live with the segment, select Members to review the results.

Example: A more complex segment

A more complex segment that includes both profile and behavioral queries gives better results that let you make a better-targeted customer journey, perhaps helping reactivate a lapsed customer. For instance, you can create a segment that includes:

  • Account name is Ford and status is active and revenue is larger than $1,000,000 (profile).

This segment would find a group of contacts who have purchased from your company in the last six months in Chicago that are now unreachable through the email address in your organizational database, suggesting that a new email needs to be established for that contact.

Phrases you can use in natural language queries

You can use a variety of phrases to find contacts and accounts with various attributes. This is not a complete list.

  • Find contacts by location

    • Contacts who live in Atlanta
    • Contacts who live in New York State
    • Contacts in Canada
    • Contacts near Seattle

  • Find contacts who share birthdays or anniversaries

    • Contacts whose birthdays are in October
    • Contacts with an anniversary in June

  • Find contacts related to accounts

    • Contacts associated with Microsoft company who live in the Czech Republic
    • Contacts associated with Microsoft company born between 1970 and 2000 who have an income greater than $50,000

  • Find contacts with upcoming life events

    • Contacts with birthdays next month

  • Find accounts missing profile information

    • Companies with no business phone

  • Find contacts with particular attributes

    • Find contacts with an email address

Tip

Customer Insights - Journeys segments are not always processed in the order that the query blocks are expressed, so it’s better to express queries in the positive “contacts who have purchased in the past 90 days” than to exclude contacts with a negative or suppression segment, like “contacts who have not purchased in the past 180 days.” Dynamic segments can be created to include, intersect, or exclude contacts in segments, allowing you to build more complex queries.