A sales manager wants Conversation Intelligence to flag whenever a competitor's product name is mentioned during a sales call, in addition to the default keyword tracking Dynamics 365 Sales provides out of the box. Reps report that competitor mentions are never highlighted in call summaries, although general sentiment and talk-ratio metrics appear correctly. What should the consultant configure?
Select an answer to reveal the explanation.
Short Explanation
Call intelligence ships with a set of things it already knows to listen for, but it isn't psychic about a specific competitor's product name unless someone tells it to listen for that term. That's done by adding the term to a custom keyword bucket in the settings; until that happens, the system simply has nothing to match against, even though its other built-in skills, like reading tone and measuring how much each person talks, keep working fine because those don't depend on a keyword list at all. The wrong paths here are chasing an audio-quality fix for a problem that isn't about transcription accuracy, inventing a separate AI Builder model that keyword tracking doesn't actually depend on, and assuming there's a special administrator role gatekeeping who can see flags. The real fix is simple: add the competitor's name to the keyword bucket.
Full Explanation
The correct answer is B. Conversation Intelligence tracks a default set of keyword categories out of the box, such as pricing or competitor mentions generally, but any organization-specific term, including a particular competitor's product name, has to be added by an administrator to a custom keyword bucket before the system will recognize and flag it in call summaries. Because sentiment and talk-ratio metrics rely on built-in models rather than configured keyword lists, they continue to work correctly even though the new competitor term was never registered. Option A is incorrect because keyword detection operates on the transcription text after it is produced; raising audio quality can improve transcription accuracy generally, but it is not the configuration step required to track a brand-new term, and nothing in the scenario points to a transcription quality problem. Option C is incorrect because there is no 'Competitor Analysis' AI Builder model required as a prerequisite for keyword-based flagging; keyword tracking is a native Conversation Intelligence configuration, not an AI Builder dependency. Option D is incorrect because keyword flags are visible to appropriately licensed sales team members and managers generally, not restricted to a separate Conversation Intelligence Administrator role, which is not a standard security role in the product.