How to Automate Claims Processing Using AI Builder and Power Automate
An insurance company wants to automate claims processing. They receive claims via email attachments (PDFs) and need to extract key information (policy number, date, amount) and enter it into a Dataverse table. Then, a workflow should evaluate the claim amount: if under $5000, auto-approve and notify the customer; if over $5000, assign to a human adjuster. The solution must be low-code and use AI where beneficial. Which combination should they use?
Quick Answer
The correct combination is AI Builder, Power Automate, and Dataverse. This works because AI Builder’s prebuilt form processing model extracts policy number, date, and amount from PDF attachments, Power Automate orchestrates the conditional workflow—auto-approving claims under $5,000 and routing larger claims to a human adjuster—and Dataverse stores the structured data in a low-code, scalable table. On the PL-900 exam, this scenario tests your understanding of how Power Platform components integrate to solve a real business problem; a common trap is choosing Power Apps for data entry when the requirement is automated extraction from emails. Remember the memory tip: “AI reads the PDF, Flow decides the path, Dataverse holds the facts.”
⚠ Common exam trap
A common mix-up: candidates think Power BI or Power Pages are needed for data handling or external access, but the scenario only requires data extraction, storage, and conditional workflow—all achievable with AI Builder, Power Automate, and Dataverse alone.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
AI Builder, Power Automate, and Dataverse
It combines AI Builder (to extract data from PDFs using prebuilt AI models), Power Automate (to orchestrate the workflow and conditional logic based on claim amount), and Dataverse (to store the extracted data in a structured table). This low-code solution leverages AI where beneficial (PDF extraction) and automates the approval routing without requiring custom development.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Power BI, AI Builder, Power Automate, and Dataverse
Why it's wrong here
Power BI renders reports and dashboards; it cannot parse PDF attachments or trigger approval workflows, so the extraction and routing steps remain unbuilt. It is tempting because AI Builder and Power Automate do appear in the stack, but Power BI contributes only visualisation, whereas the scenario needs Power Automate alone to orchestrate the email intake.
- ✓
AI Builder, Power Automate, and Dataverse
Why this is correct
AI Builder's document-processing model extracts policy number, date and amount from the PDFs; Power Automate orchestrates the trigger, conditional branch on the $5,000 threshold, approvals and notifications; Dataverse stores the structured claim records. All are low-code, meeting the stated constraint.
- ✗
Power Pages, AI Builder, Power Automate, and Dataverse
Why it's wrong here
Power Pages builds external-facing websites and portals; it cannot receive mailbox attachments or run the amount-based approval branch. It is tempting because it is a low-code maker tool that pairs with Dataverse, but the scenario's entry point is an email inbox, which Power Automate handles, not a web portal.
- ✗
AI Builder, Copilot Studio, Power Automate, and Dataverse
Why it's wrong here
Copilot Studio builds conversational agents and chatbots, not the automated email-attachment intake and conditional approval routing this scenario requires. It is tempting because it is a low-code AI tool that sits alongside Power Automate, yet the claim must be triggered by an incoming email rather than a user conversation, so the orchestration layer is missing.
Go deeper
Related to this question
About these practice questions
This PL-900 question is part of Courseiva's 701-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
Same concept, more angles
1 more way this is tested on PL-900
These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.
Variation 1. An organization uses Power Automate to process invoices from email attachments. Some invoices are processed incorrectly because the flow fails to parse PDFs with non-standard layouts. What should the administrator do to improve accuracy?
hard- A.Increase the flow run frequency to every minute.
- B.Add a condition to route non-parsed invoices for manual approval.
- ✓ C.Use AI Builder to create a custom document processing model.
- D.Change the trigger to a manual button in Power Apps.
Why C: AI Builder provides prebuilt and custom document processing models that use optical character recognition (OCR) and machine learning to extract data from PDFs with non-standard layouts. This directly addresses the root cause—parsing failures due to layout variability—by training the model on sample invoices to handle diverse formats, unlike static parsing logic.
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PL-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the PL-900 exam.