What if your Dataverse tables could think for themselves?

With the new Prompt field capability in Microsoft Dataverse, we’re entering a new era where business applications don’t just store data—they understand it.

📌 What Is the Prompt Feature in Dataverse? In simple terms, this feature lets you embed a natural language AI prompt directly into a Dataverse column. It leverages Microsoft Copilot Studio (built on Azure OpenAI) to analyze other data in the record and auto-generate content or insights into that column.

You define:

• What you want the AI to do (e.g., analyze, summarize, generate)
• The context fields (other columns) to draw from
• The format or structure of the response

In the example shown above, the Sentiment column has a prompt that reads the Review Text and returns a value like Positive, Negative, or Neutral—automatically.

No need to build and call custom AI models. No extra automation or plugin. It’s all happening inside Dataverse.

🔍 Real-World Business Scenarios

Let’s explore how this can be transformative across industries:

1. Customer Feedback Sentiment Analysis

Automatically analyze thousands of customer reviews, complaints, or survey responses to:

• Identify satisfaction trends
• Flag negative feedback for escalation
• Feed Power BI sentiment dashboards

2. Sales Opportunity Qualification

In a CRM system, use a prompt field to summarize call notes and return a lead rating or intent score:

“Based on the meeting notes, rate this lead as Hot, Warm, or Cold.”

3. HR Screening & Resume Summaries

Parse open-ended responses from job applications or resumes to:

• Summarize qualifications
• Identify red flags or key experience
• Pre-fill tags or shortlist scores

4. Risk Classification in Insurance or Legal Cases

For records describing claims or incidents, a prompt can:

• Classify risk level
• Extract liability indicators
• Recommend next actions

5. Product Description Enhancement

Enhance or rewrite brief product notes into full marketing-friendly descriptions using a prompt that considers key features and tone preferences.

✨ Why This Matters

Traditionally, adding intelligence like this required:

• Custom models or integration with Azure Cognitive Services
• Power Automate flows to call APIs
• Complex setup and monitoring

Now, with prompts embedded at the column level, it’s:

• Low-code
• Instant
• Contextual (it’s aware of the record it’s in)

It empowers business users—not just developers—to design smarter tables and make data more actionable from the source.

⚙️ How to Use It

1. Create or edit a column in your Dataverse table.
2. Set the data type as “Text” (prompt only works with Text for now).
3. Enable “Allow form fill assistance (preview)” and define your prompt.
4. Reference other columns using dynamic tokens.
5. Save—and watch your data come alive!

🧠 Pro Tip: Keep prompts short and clear. Focus on the outcome you want (e.g., classify, summarize, generate) and test with sample records to fine-tune tone and accuracy.

📈 Final Thoughts

Prompt-enabled columns in Dataverse mark a subtle but significant shift: AI becomes part of your data model, not just layered on top of it.

It’s not about replacing humans—it’s about accelerating insight, reducing manual effort, and scaling intelligence across your business applications.