AB-210 Accelerating Sales Pipelines with AI in Dynamics 365 Question and Answers

Requirement

Agent mode

Recommend action for disengaged deals

Research

Generate insights from records and research into a unified deal view

Research

Ensure customer interactions are compliant

Engage

The first two requirements belong to Research mode because they are insight-generation functions for opportunity management. Microsoft describes the opportunity-focused Sales agent as identifying risks such as disengaged stakeholders, slowing momentum, and competitive pressure, then recommending targeted actions to keep the deal moving forward. It also provides a comprehensive deal view by surfacing buyer needs, stakeholder roles, competitor mentions, and deal health in one unified place. This directly matches both “recommend targeted action for disengaged deals” and “generate insights from CRM records, emails, meetings, and web research into a unified deal view.”

The compliance requirement maps to Engage mode because compliance matters when the agent communicates externally with customers. Sales Close Agent Engage mode automates customer engagement through outreach and follow-up emails, and Microsoft states that admins configure an email delivery compliance profile so the agent’s communications follow organizational policies and regulatory requirements. Research mode produces insights for sellers; it does not perform customer outreach. Therefore, use Research for opportunity intelligence and Engage for compliant customer interaction.

References/topics: Sales Close Agent; Sales Opportunity Agent Research mode; Sales Close Agent Engage mode; opportunity risk insights; compliance profile for agent email communications.

Requirement

Automation type

Notify a manager when a high-value opportunity is created

Power Automate automated cloud flow

Validate the products with an external ERP system every Sunday

Power Automate scheduled cloud flow

Display a guided lead qualification interface on the lead form

Business process flow

Update an external ERP system when an opportunity is won

Power Automate automated cloud flow

A Power Automate automated cloud flow is the correct automation type when work must run automatically after a business event occurs. Microsoft describes cloud flows as flows that can perform actions automatically after an event triggers them, such as sending a notification after a triggering event occurs. Therefore, notifying a manager when a high-value opportunity is created and updating an external ERP system when an opportunity is won are both event-driven scenarios, so they should use automated cloud flows with Dataverse triggers.

A Power Automate scheduled cloud flow is correct for validating products with an external ERP system every Sunday. Microsoft states that a scheduled cloud flow runs at a specific time or recurring schedule, such as daily or weekly, and uses a recurrence trigger. That directly matches the weekly Sunday validation requirement.

A Business process flow is correct for displaying a guided lead qualification interface on the lead form. Business process flows provide a streamlined user experience that guides users through defined processes, and Dynamics 365 Sales documentation explains that stages and steps display in a control at the top of a form. Instant cloud flow is not used because none of the requirements ask for a manually triggered user action.

References/topics: Power Automate automated cloud flows; scheduled cloud flows; Dataverse triggers; Business process flows; Dynamics 365 Sales automation and extensibility.

The correct agent type is Sales Qualification Agent: Research and engage . The requirement is specifically about inbound lead processing: researching incoming leads, generating outreach, autonomously contacting leads, following up, and evaluating qualification during the engagement process. Microsoft defines Sales Qualification Agent as the AI agent used to automate lead qualification in Dynamics 365 Sales, especially for high-volume inbound leads from sources such as websites, events, and webinars. In Research and engage mode, the agent researches the lead, evaluates fit using target customer profile and BANT criteria, sends personalized outreach emails, engages based on responses, sends follow-up emails, detects positive intent, and then hands promising leads to sellers.

Research-only mode is not enough because it generates a draft outreach email for seller review but does not send outreach emails or detect positive intent from responses. Microsoft’s mode comparison explicitly shows that BANT checking, sending outreach, detecting positive intent, and follow-up emails are available in Research and engage mode, not Research-only mode. Sales Close Agent options are wrong because they apply to opportunity/deal execution, not initial lead qualification.

References/topics: Sales Qualification Agent; Research and engage mode; lead research; outreach automation; BANT qualification; AI-driven lead prioritization.

The correct insight to share is Recommended actions . Sales Close Agent in Research mode is opportunity-focused: it helps sellers understand the state of a deal, identify risks, and decide what to do next. Microsoft explains that the opportunity agent provides critical insights and recommendations so sellers can ramp up on opportunities, engage the right stakeholders, and mitigate risks effectively. The Responsible AI FAQ also describes the agent as identifying deal risks early and providing recommended mitigation strategies based on historical data and communication analysis.

The question specifically says the seller must decide how to proceed and wants to move the deal forward. That points directly to recommended actions, not static context. Financial health may be part of a broader opportunity view, but it is not the best answer for next-step guidance. Budget, Authority, Need, and Timeline is BANT qualification and belongs to lead qualification scenarios, especially Sales Qualification Agent, not opportunity research. Blueprints are generated by the Sales Research Agent on a research canvas for analytics-style exploration, not by the Sales Close Agent opportunity summary. Therefore, the correct insight category is Recommended actions .

References/topics: Sales Close Agent Research mode; opportunity recommendations; deal risk mitigation; Sales Opportunity Agent insights; opportunity progression.

The issue is that the sales manager already generated a Sales Research Agent blueprint but forgot to include the quarterly competitive intelligence brief. The brief is an external document and the case study states it is delivered as a PDF document , so the correct first content action is to upload the competitive intelligence brief as a PDF file , not CSV. Microsoft’s Sales Research Agent documentation states that users can select Add data to upload files including CSV, Excel, or PDF to augment research with additional data.

The second required action is to add a data source , because the competitive intelligence brief is not part of the default Dynamics 365 Sales data. Sales Research Agent connects to Dynamics 365 Sales data by default, but additional external files must be added as supporting data sources before they can influence the analysis. The documentation identifies the Data area as the place where the research data source is specified and where additional data sources can be added.

The final action is to regenerate the canvas/blueprint so the agent reprocesses the analysis using the newly added PDF. Refreshing alone is not the best answer because the problem is not stale data; it is missing context. Regeneration creates updated research output with the new source included.

References/topics: Sales Research Agent; research canvas; adding PDF data sources; blueprint regeneration; intelligent opportunity research.

The correct configuration is to set related record creation to Seller. The case study states that sales leadership requires flexibility in how leads are converted: inside sales need automatic opportunity creation, but field sellers need the option to manually decide whether an opportunity should be created when a lead is qualified. In the new lead qualification experience, Dynamics 365 Sales lets administrators control whether account, contact, and opportunity records are created automatically during lead qualification or whether sellers are prompted to choose which related records to create. Microsoft’s lead qualification guidance confirms that when automatic creation is not enabled, sellers are prompted to select which records should be created during qualification.

Option B and option E are wrong for field sellers because automatic creation would remove the manual decision point required by the case study. Option C is relevant to guiding sellers through sales stages, but the existing Lead to Opportunity Business Process Flow does not control whether related records are automatically created when a lead is qualified. Option D is unrelated because Copilot summaries help sellers review record information; they do not configure lead conversion behavior. Therefore, the required configuration for field sellers is Seller-controlled related record creation.

References/topics: Lead qualification experience; lead-to-opportunity conversion; related account, contact, and opportunity creation; Dynamics 365 Sales lead management.

Create a Power Automate cloud flow.

Add a Dataverse trigger.

Define an approval step.

Test and activate the flow.

The correct automation pattern is an automated Power Automate cloud flow triggered by a Dataverse event. The case study states that the process must notify the seller’s manager when a draft quote is created and requires approval, and that approvals must use the Power Automate Approvals connector. Therefore, the flow should first be created as a cloud flow, then started by a Dataverse trigger such as when a row is added, modified, or deleted on the Quote table, filtered for the draft quote condition. Microsoft’s Dataverse trigger guidance confirms that this trigger runs when a selected Dataverse table row is created, modified, or deleted.

After the trigger is defined, the flow must include an approval step using Power Automate approvals. Microsoft’s approval workflow guidance states that Power Automate can manage approvals across services including Dynamics 365 and Dataverse-based processes. The final action is to test and activate the flow so the quote approval automation is available to users. Do not use a scheduled trigger because the requirement is event-driven, not time-based. Do not connect to the existing desktop flow because that desktop flow processes partner orders for fulfillment, not draft quote approvals. A Teams step is optional collaboration, not the required approval mechanism.

References/topics: Power Automate cloud flows; Dataverse triggers; Approvals connector; Dynamics 365 Sales quote approval automation; sales platform extension.

Requirement

Tool

Perform purchase required by CFO

Microsoft 365 admin center

Configure Sandbox billing

Power Platform admin center

Configure Production billing

Microsoft 365 admin center

Why: Production requires reliable AI capacity so sellers are not blocked when prepaid Copilot capacity is depleted. The correct production setup is pay-as-you-go / billing policy configuration through the Microsoft 365 admin center . Azure portal is not the best answer here because the scenario says no Azure subscriptions are available , and Microsoft 365 admin center can be used for the billing policy workflow. Microsoft Agent 365 is for agent management/governance, not billing configuration.

For Dynamics 365 Sales predictive lead scoring, the model cannot be created unless there is enough historical lead outcome data in the selected training window. Microsoft’s predictive lead scoring configuration requirement states that the organization must have at least 40 qualified leads and 40 disqualified leads in the selected training period. The official guidance specifically explains that if the model is trained with leads from the past three months, the organization needs “at least 40 qualified and 40 disqualified leads” created and closed during that period.

In the sandbox, AdventureWorks has 30 qualified leads and 20 disqualified leads. However, 10 of the qualified leads are older than three months, and the business requirement says predictive scoring must be enabled for leads created within the last three months. Therefore, only 20 qualified leads are usable for this model. The sandbox currently has 20 usable qualified leads and 20 disqualified leads. To meet the required threshold of 40 each, the company needs 20 more qualified leads and 20 more disqualified leads . Simply creating 40 new leads is not enough because predictive scoring requires outcome-classified leads, not merely open or newly created records.

References/topics: Configure predictive lead scoring; lead scoring model training data requirements; AI-driven lead prioritization in Dynamics 365 Sales.

Requirement

Component

Mountaineer range

Product family

Athlete package

Product bundle

Metal range

Product property

The Mountaineer range should be configured as a Product family because it represents a product classification hierarchy. Dynamics 365 Sales uses product families to organize related products, child products, and bundles under a parent structure. Microsoft’s product catalog guidance states that product families can contain child product families, products, and bundles in a hierarchy, which fits the mountaineer cycle range with economy and premium fitting variations.

The Athlete package should be configured as a Product bundle because it combines multiple sellable items: a cycle, connected devices, and subscription-based analytics services. A bundle is used when multiple products are sold together as one commercial package. Dynamics 365 Sales allows products to be associated with a bundle, including quantities and required or optional components.

The Metal range should be configured as a Product property because sellers must choose a configurable attribute—chrome, aluminum, or titanium—when adding the mountaineer cycle to an opportunity. Product properties define product characteristics such as size, color, or component, and can be used at run time when adding products to opportunities, quotes, orders, or invoices.

References/topics: Product catalog setup; product families; product bundles; product properties; opportunity product configuration.

The required fix is about visibility and security , not agent operating mode. In the scenario, production has EnableRoleBasedSystemViews set to true, which restricts system views based on role configuration. Because the business requirement says all sellers should be able to access disqualified leads , the role-based system view setting must be updated so sellers can access the view that exposes leads disqualified by the Sales Qualification Agent.

The second required action is to change Lead table permissions. The customized Salesperson role currently grants sellers read access only to leads they own. That is too restrictive because leads disqualified by the Sales Qualification Agent may not be owned by each seller. To let all sellers view those records, sellers need organization-level Read permission on the Lead table. Option A is wrong because upgrading the agent to research and engage mode affects how the agent researches and interacts with leads; it does not solve seller access to disqualified records. Option C is also wrong because the look-back period controls which historical leads the agent evaluates, not who can view disqualified leads.

References/topics: Sales Qualification Agent; role-based system views; Dataverse security roles; Lead table read access; AI-driven lead qualification.