AI assisted query builder

Important

This content is archived and is not being updated. For the latest documentation, see Microsoft Dynamics 365 product documentation. For the latest release plans, see Dynamics 365 and Microsoft Power Platform release plans.

Note

These release notes describe functionality that may not have been released yet. To see when this functionality is planned to release, please review What’s new and planned for Market Insights. Delivery timelines and projected functionality may change or may not ship (see Microsoft policy).

To get relevant market insights, it is extremely important to ensure that search rules are set up correctly to provide insights on brands, products, and topics that are very relevant and precise to the user’s needs. Currently, it can be difficult for users to get the configuration of search rules correct on the first run because it typically takes weeks of iterations before the rules are refined to the level of precision required.

A few reasons make this difficult:

• Consumers search for and mention brands in many ways (for example, Chevy for Chevrolet, FB for Facebook).
• Brand names can conflict and refer to entities in more than one industry (for example, Dove is a soap and a chocolate brand).
• Users often want to exclude specific entities from their topics (for example, a user might want to track trends on “Nike Air Jordan 1” but not the “High Zip” model).
• Users can’t always anticipate what related topics are worth monitoring (for example, consumers who searched for Nike Air searched for what else?).

As a result, the relevance and cost of generating insights heavily depend on cutting out the noise and getting your search set up right.

Our AI-powered assisted query builder solves this core problem. As users start typing their search term, they will be presented with suggested terms most relevant to them. For example, if a user types “Eag...,” they can choose from “Eagle” (the bird), “Philadelphia Eagles” (the football team), “Eagle Scout” (the Boy Scouts), and more. By providing users with a highly relevant list of brands, products, and entities to select, the assisted query builder helps to reduce the cognitive load of search setup so that they spend less time setting up their searches and more time acting on the insights.