<img src="https://secure.leadforensics.com/133892.png" alt="" style="display:none;">

Back in April I attended the Power BI Summit which had more than 100 presentations, and I came away with the conviction that we have come a long way in the Power BI world.
One thing clearly stood out in the flurry of news and presentations: namely, how Microsoft has really made great strides to integrate AI technology into Power BI Pro & Premium.
Until recently, most businesses assumed you needed a Ph.D. in computer science to work with AI. But platform providers like Microsoft have been busy integrating AI tools into their products and services in order to put that power into the hands of ordinary users and developers.
It's pretty wild what the combination of Power BI and AI lets you do today!

Let’s look at just 5 of the new AI “hacks” available in Power BI, which will really make your BI products pop.

Power BI AI hack # 1: Automated Analyses

Good reporting is about telling a story, and data reporting is no different. It can be the story of the result of a campaign, or the story of the development in earnings over the last 3 years. The art is to identify the real story, and then convey it.

With Power BI you have an alternative to starting the analysis from scratch, namely the “Smart Narratives” feature. When you choose the Smart Narratives visualization Power BI adds a box to your report with an analysis of the data already written for you, noting important trends and data points. From here you can then start diving into data and researching different dimensions, adding your own insights and understanding as you edit the text.
Smart Narratives is a shortcut to building reports, but it is also an opportunity to gain completely new insights. At the same time we can reduce the risk of “confirmation bias,” where we often find only the results and explanations that we are looking for, consciously or unconsciously.

In addition to Smart Narratives, Power BI also has a Q&A function where you can enter questions about the data and Power BI provides answers. Power BI can learn to understand what you are asking about in your own words. If you use the word 'school', for example, Power BI may not understand it, but you could then link it to 'education' - and Power BI then learns that 'school' is equal to 'education'.

Power BI AI hack # 2: Anomalies

When you work with time series data, such as earnings over time, there will naturally be variations in the data from week to week or month to month. Traditionally analysts would study a graph reminiscent of a jagged mountain ridge and have to assess whether everything looks normal or if there are important surprises in the data.

Large fluctuations in the upward or downward direction are usually visually obvious. But does this mean that there is an irregularity in the underlying data, or is it an expected fluctuation- for example because you have run a new sales campaign?

These are important distinctions, but generally easy to diagnose. It is more difficult to spot the 'invisible' irregularities - where there is nothing that stands out to the human eye, perhaps because the data is highly volatile.

With the Analytics function 'Anomalies', Power BI uses AI to look at the underlying data and automatically detects if there are irregularities in your time series data. The Anomalies feature also provides explanations of the irregularities to help with root cause analysis.

Of course, each time series is different and the threshold for what counts as an important anomaly can vary a lot. Therefore, you can control how sensitive the algorithm in Power BI should be to changes in data, and when a deviation should be marked as an irregularity.

Power BI AI hack # 3: Analyze

You have reported the turnover figures for the current year to the sales director in a clear bar chart, month by month. The month of February shows a marked increase in revenue - and now the director would like an explanation for the increase.

Instead of having to embark on a laborious detective job, right-click on the column for February. In no time, you have a new analysis of the February figures, clearly presented. Here you can see that it was Denmark, Germany and the USA in particular that experienced an increase in revenue in February, while Denmark, Germany and France had the largest relative change in sales figures.

With this AI functionality, you move from getting answers to what's happening, to also getting answers to why it's happening. AI looks deeper into the numbers, and looks behind patterns and trends. In the example here, we have taken as our starting point a finished report, which the sales director wants to elaborate on - but it is also obvious to use the Analyze functionality in connection with the report being built up.

Power BI AI hack # 4: Decomposition Tree

Another way to dive behind the numbers is by using Power BI's functionality “Decomposition Tree.”

The thinking behind the Decomposition Tree is: What value would you like to analyze (a measurement or aggregation of data), and what dimensions would you like to explain the value by (by performing data derivation)? Maybe you want to look at your supply chain and analyze what percentage of your products you have in backorder? Or maybe you want to analyze sales of a particular product category (books) by various factors, such as genre and publisher?

With the Decomposition Tree function in Power BI, you can visualize data across multiple dimensions. The function automatically collects data, so you can perform data extraction in your dimensions in any order. You can use the tool for ad hoc exploration and root cause analysis - and you can also ask Power BI to find the next dimension in your data so that you can perform detailed deductions based on specific criteria.

Power BI AI hack # 5: Visualization of key influencers 

Data reporting is about the “what” but data analysis is always about the “why:”

- Your customer service manager wants to know why your customers are giving your call center bad reviews.

- Your product manager wants to know, what are the characteristics of your product that customers are particularly happy with, and what they are especially tired of?

- Your sales manager would like to know if these are the same factors that cause sales to increase in the Danish and Norwegian markets?

- And your HR manager has a gut feeling that commuting time and seniority are important for the increase in employee turnover - but is that right, and which weighs heaviest?

The visualization of Key Influencers helps you understand the factors behind the value you are interested in. Key Influencers analyze data, rank the factors that matter, and show them as key factors behind your outcomes of interest.

With this tool you can see not just the factors that affect, for example, employee turnover, but also compare the relative importance of the factors and see if, for example, commuting time has a greater impact on employee turnover than seniority.


Be sure to explore these and other new features available in Power BI on your own data and see how much more your BI pops with a little AI on top!

I hope that this blog found you well, and feel free to get in touch if you have any questions.


Discuss this post

Recommended posts

As artificial intelligence (AI) and machine learning (ML) reshape industries and redefine possibilities, mastering the art of AI and ML model development and evaluation has never been more crucial. Behind every powerful AI or ML model lies strategic decisions, intricate algorithms and critical evaluations. Within the Model and Evaluate phases of the Columbus AI Innovation Lab, we delve into these aspects and how to use diverse metrics to evaluate a machine learning model’s performance, along with its strengths and weaknesses.
It doesn’t matter how good your artificial intelligence technology is. If you have bad data, you’ll get bad results. While AI is one of the most innovative and impactful innovations for businesses today, bad data can prevent AI projects from getting off the ground. Having high-quality data and following data preparation best practices are crucial for a successful AI project. In the Transform phase of the Columbus AI Innovation Lab, the one we detail here, data is collected and prepared for use by AI and machine learning (ML) algorithms. Major phases in Machine Learning Machine learning, frequently used synonymously with “artificial intelligence,” is about using predictive methods to simplify tasks. There are several essential steps to building ML models, and these steps can change based on the issue you are trying to solve and the kind of ML model you are building. However, in general, the steps involved in creating and implementing an ML model are as follows: Define: Different approaches for finding AI use cases Discover: Scope and prioritize ideas before developing an AI solution and implementation strategy Transform: Transforming the business case including data acquisition and preparation with AI/ML (current blog) Model and evaluate: Analyze and enhance the business use cases using AI/ML algorithms Deploy and support: Deployment of AI/ML models and support/monitoring to guarantee quality and effectiveness
Welcome to the exhilarating world of the AI revolution! In this blog series we will dive deep into the most relevant and fascinating topics to help you achieve great results with AI for your business. Lately, AI has been dominating the headlines, creating a whirlwind of buzz that's hard to keep up with. In this first blog we are giving a quick insight to the current status!
Only half of the companies starting an AI pilot project are actually executing it. The key is to choose an idea that will benefit your business. Read more about how! In 2022, 27% of chief information officers confirmed they deployed artificial intelligence (AI), according to a Gartner AI survey. Even though businesses across all industries are turning to AI and machine learning, prepare your organization before jumping on the AI bandwagon by considering a few factors. Ask yourself: Is AI necessary for achieving the project requirements or is there another way? Does your team have the skills to support AI and machine learning? How will AI impact your current operations if you adopt it? How will you integrate AI with existing systems? What are the data, security and infrastructure requirements of AI and machine learning? The Gartner AI survey found only 54% of projects made it from the pilot phase to production. After significant investment in AI, why aren’t companies deploying it? We found the problem begins when companies define a use case. Too often, companies are not identifying AI use cases that benefit their businesses and end-users will adopt. The question is then, how should companies unlock the value and new opportunities AI promises? It starts with a systematic approach for each stage of the AI life cycle. We developed the Columbus AI Innovation Lab, a comprehensive method to address and account for all challenges when adding AI to your business operations and bring stakeholders into the process at the right time to help you operationalize AI.
If you have identified possible AI use cases for your business, the next step will be to test if they are possible to implement and if they will create great value. While there is a lot of momentum and excitement about using AI to propel your business, the reality is only 54% of AI projects are deployed. How do you ensure you’re one of the businesses that does unlock the new opportunities AI promises? Your success with AI begins by discovering AI use cases that work for your business. In the first blog of our Columbus AI blog series, we shared five areas where organizations should focus their efforts to generate ideas for AI implementations based on our experience. After generating some ideas for AI use cases that could potentially benefit your company from the first step of the Columbus AI Innovation Lab, the next step is to test which AI use cases could be operationalized by evaluating them. Columbus AI Innovation Lab
right-arrow share search phone phone-filled menu filter envelope envelope-filled close checkmark caret-down arrow-up arrow-right arrow-left arrow-down