Author: In today’s fast-paced, data-driven environment, BI developers are constantly looking for ways to shorten development cycles, improve report quality, and deliver insights at scale. Here, we share best practices on ways for visualization teams to leverage AI in data analytics to scale the design and development of BI solutions.
With the evolution of AI tools, BI developers now have intelligent assistants that automate repetitive tasks, enhance accuracy, and significantly boost productivity across the reporting lifecycle.
At CoStrategix, our Visualization Engineering team conducted several AI-first experiments to evaluate productivity gains of using AI in data analytics, focusing on the areas where developers spend the most time.
This blog presents three practical, tested automation use cases that significantly reduce reporting lifecycle effort, while maintaining quality and developer control.
1. Creating a Theme Template with AI
Creating visually cohesive Power BI themes across multi-page dashboards is often a time-intensive task. Traditionally, developers spend hours manually defining JSON theme files, and adjusting fonts, headers, and colors for each visual.
CoStrategix tested both Copilot and Gemini to automate the creation of theme templates. We found that Gemini outperformed on theme creation, and our team has now automated this process entirely. The AI-generated theme files are comprehensive and production-ready, with detailed specifications for typography, colors, and visual hierarchy.
Outcome
We achieved a 95% time savings in manually formatting the visuals.
How to Automate Theme Template Production with AI
Step 1: Collect Brand Metadata
- Primary/secondary accent data colors
- Typography guidelines
- Report background and foreground colors
- Table accent colors (grid lines, headers, selection highlights)
Step 2: Prompt Gemini to generate theme JSON via persona-based prompts
Step 3: Deploy + Validate in Your BI Tool
- Import JSON → run UI smoke tests
- Validate accessibility
- Confirm visual hierarchy
2. Accelerating Data Model Development with AI
At the heart of every BI solution lies an effective data model. Building it requires identifying fact and dimension tables, defining relationships, and ensuring the right filter propagation tasks, which can be both repetitive and error-prone. While we typically build a star schema in the medallion architecture, it could be done in the BI in certain cases, so we tested that.
We tested both Copilot and Gemini to automatically analyze table structures and generate recommendations for star schema models, including relationship definitions and cardinality settings. We found that Copilot performed better in this task, reducing not only design time but also helping to standardize modeling best practices across developers.
Outcome
While Copilot’s suggestions still required human validation, Copilot acted as a skilled assistant in laying the foundation quickly and accurately, so developers could focus on optimization and validation.
We averaged about a 92% time savings in analyzing and building the Data Model with AI assistance. Copilot helps in automatic schema interpretation and auto-generated relationships.
How to Accelerate Data Model Development with AI
Step 1: Provide Source Metadata
- Table schemas, column names, sample rows, foreign keys, ER diagrams
Step 2: Leverage Copilot
- Auto-detect relationships and cardinality
- Suggest transformations
Step 3: Validate and Build an Efficient Data Model
- Refine the star schema
- Check cross-filter directions
- Validate relationship logic
3. Simplifying Complex DAX Calculations with AI
Writing accurate Data Analysis Expressions (DAX) measures is often the most intellectually demanding part of BI development. From handling disconnected tables to building composite columns, even experienced developers can spend hours refining logic.
CoStrategix tested both Copilot and Gemini to generate precise, well-documented DAX formulas for complex analytical scenarios. We found that Gemini outperformed in this task. It also provided natural language explanations for each function, making it easier for developers to understand and validate the results. This demonstrates Gemini’s potential as a knowledge accelerator, not just automating code generation. It also enables deeper learning and understanding within our team.
Outcome
While Gemini’s suggestions still required human validation, Gemini acted as a skilled assistant in laying the foundation quickly and accurately, so our developers could focus on optimization and validation.
Our team achieved a 95% time savings in creating a very complex DAX measure using AI, producing a validated DAX measure with test cases and performance notes.
How to Simplify Complex DAX Calculations with AI
Step 1: Describe the business logic in natural language. For example: Handling exceptions, fiscal calendars, disconnected slicers
Step 2: Use a personal prompt to generate performance-efficient DAX code
Step 3: Validate the DAX code and deploy
Overall Observations About AI Tools
Based on our exploration, the strengths of each AI tool are becoming clearer:
Copilot
- Excels in structural automation
- Data model analysis
- Schema and relationship recommendations
- Transformation code generation
Gemini
- Excels in creative and analytical automation
- Theme creation
- Complex DAX logic
- Detailed explanation and reasoning
Together, they complement each other across different stages of BI development.
As CoStrategix continues to act as an AI-first data organization, these experiments reinforce that AI will be central to the future of business intelligence. Instead of replacing developers, AI tools in data analytics are amplifying developers’ capabilities – helping them to work smarter, not harder.
With Copilot and Gemini integrated into our workflows, CoStrategix developers are spending less time on manual, repetitive tasks and more time crafting insights, designing better user experiences, and delivering high-impact analytics that drive business value.
The way forward is clear. BI’s next era will be shaped by human expertise and AI working together to create analytics that drive real business value.
CoStrategix is a strategic technology consulting and implementation company that bridges the gap between technology and business teams to build value with digital and data solutions. If you are looking for guidance on data management strategies and how to mature your data analytics capabilities, we can help you leverage best practices to enhance the value of your data. Get in touch!
AI Strategy & Solutions – Elevate your business with advanced analytics
Data & Insights – Drive insights that lead to competitive advantage
Product Development – Build platforms that power unique digital capabilities
Platform & Technology Modernization – Modernize for stellar experiences, efficiency, and AI
Related Blog Posts
Measuring the Value of Your AI Initiatives
November 20, 2025
Transcend the AI Hype with a Product Mindset
November 17, 2025
Mastering Data Quality
November 7, 2025
Model Context Protocol: A New Standard for AI Integration
September 22, 2025
