top of page
Search
Writer's pictureCarlos Altamirano
Sep 173 min read

Smart Data Analytics: Power BI and ChatGPT in Action | Xyclos

Updated: Sep 19


The advantages of using Power BI and ChatGPT together for data analysis | Xyclos




The best of both worlds.


In Power BI, I created 3 measures to understand the performance of a client's bottle production:


  1. Total bottles produced

  2. % of broken bottles

  3. % of broken bottles


Use the Stacked Column and Line Chart visual to graph all 3 measures in a single chart


Then I go to ChatGPT to do the analysis for me, entering the prompt along with the graph copied from Power BI and this is the result of the analysis:


I have this graph (visual object), please analyze it and give me a summary of the most relevant things you observe or what is happening.


The graph you've provided seems to be a combination of a bar and line chart, displaying several metrics related to the production or distribution of bottles, such as total bottle output, broken bottles, and shattered bottles. Here is an analysis of the key observations:


  1. Total Bottle Output (Bar Chart - Orange):

    • The total output seems to increase and peak in April (2.27M) and May (2.32M). After this peak, there's a steady decline starting from June (1.95M) and continuing into August (1.55M).

    • The lowest value is in February (0.89M).

  2. Percentage of Broken Bottles (Line - Green):

    • The percentage of broken bottles increases steadily between January (0.98%) and March (1.63%), peaking in April (2.31%).

    • There is a significant drop in this percentage from May (2.22%) to August (1.13%), which suggests improvement in reducing the rate of broken bottles over time.

  3. Percentage of Shattered Bottles (Line - Dark Green):

    • The percentage of shattered bottles seems to follow a downward trend after April (1.00%), hitting its lowest point in August (0.44%).

    • The graph shows a steady improvement, as the percentage of shattered bottles decreases significantly from the earlier months.

Key Takeaways:

  • Production peaks in April and May but begins to decline in the following months.

  • Both broken and shattered bottles peak around April and then decrease steadily, suggesting that corrective measures may have been put in place to reduce the number of defective bottles after the high percentages in the first half of the year.

  • The most significant improvement in quality (i.e., fewer broken or shattered bottles) happens from April to August.


This chart appears to highlight the relationship between bottle production and defect rates, with notable efforts to improve quality control in the latter part of the year.


Next step? Prepare to work safely and confidently with Power BI


In this case, it was relatively obvious what was happening, but it didn't hurt to ask "someone else" what they thought, so I asked ChatGPT for their opinion.


Using advanced tools like Power BI and ChatGPT together opens up a world of possibilities to improve and optimize data analysis.


Creating custom measures in Power BI and selecting the right visuals, such as stacked column and line charts, are essential skills for generating clear and accurate reports.


By incorporating ChatGPT to support your analysis, you not only save time, but you gain an objective and in-depth perspective that can identify patterns and key points that might otherwise go unnoticed.


This highlights the importance of mastering tools such as Power BI , as it enables us to perform effective visual analysis, while AI support adds a layer of intelligence that improves data interpretation, facilitating more informed decisions and stronger strategies in any area of analysis.



In this Power BI course, you will learn how to create Data Models, use DAX in depth to create Measures, Calculated Columns; how to efficiently program your measures and organize measures among other important topics that we cover in this course.


But we don't just go over concepts, you'll put into practice everything you learn to ensure that while you prepare, you gain real, demonstrable experience using Power BI.



8 views0 comments

Comments

bottom of page