Augmented BI Use Cases
Augmented business intelligence will transform how we monitor and analyze business performance, truly enabling data-driven culture to thrive in your organization.
In my previous blog, we looked at how our augmented business intelligence (BI) tool leverages machine learning technology to automate time-consuming tasks and put advanced analytical capability into the hands of business users. When partnered with SAP Data Warehouse Cloud, augmented BI eliminates unneeded manual tasks, allows business analysts to dig deeper and achieve a better understanding of the business, and frees up scarce data science resources for more advanced, higher-value projects—all with the confidence that comes from knowing that your data is current, complete, and secure in an environment the solves the complexity of data sprawl and multiple user types.
Now let’s take a look at three use cases where the advantages of augmented BI over a traditional BI toolkit become clear:
- Business Monitoring
- Data Exploration
- Data-Driven Decisions
In traditional BI solutions, dashboards and alerting are common tools used to monitor for KPI changes. Users monitor in search of problems they need to be aware of, then when a problem is found, they have to dig in further to understand what it is happening. A good example would be a failure to meet a defined target for sales at the end of the month.
Augmented business intelligence will take the concept of monitoring to the next level. Instead of identifying problems, the focus can be on alerting users that there is a change and understanding why it is happening, regardless of whether the target is met. For example, if your pipeline should be at $100 million by end of March, and you are on course to hit that number, you would still want to know if the pipeline for China is 20% below where it was at this time last year—and that that drop was caused by retention problems in the sales force. You would want to know because the drop in Chinese pipeline could raise problems downstream, or maybe this is an indication that your target should be more than $100 million.
The standard problem-reporting approach to business intelligence would have you leaving money on the table or discovering the problem too late. But the augmented BI approach could automatically alert you that there is a change that is worth understanding. The system can learn the priorities of your business both by your behavior—if you spend time on the dashboard looking at your pipeline every day, it’s clearly something very important to you—and by your stated preferences. The system can look for trends within historical values that may be significant, while sparing you from having to look across hundreds or thousands of dimensions to figure out what may be important.
In traditional BI environments, people devote a major share of their time to building dashboards and visualizing their data. It is a lot of time spent on very meticulous manual processes.
Of course, the real goal isn’t figuring out how to visualize the data. The goal is to find the information needed and share it with others. Beginning with that understanding, in an augmented Business intelligence environment, with a small amount of guidance the system identifies what information you should highlight and automatically creates a beautifully formatted dashboard or a story for you.
If, for example, you have been looking at employee retention, the system will break down the distribution of the amount of days that individuals have been with the company, the total number of people turning over, hotspots by country or by business area, and other relevant factors. Via embedded machine learning, it will automatically identify outliers and suggest relationships within the data that may be significant.
With all that work done automatically, you can then spend your time tweaking the presentation for a particular audience or adding information that helps to complete the story—but you are already 80% of the way there. Augmented BI accelerates what you need to do, focusing you on the higher-level questions that you and your stakeholders need answers for, rather than the nitty-gritty of adding a chart, choosing a chart type, choosing the data, etc.
“Data driven decision-making” is a catchphrase that we have used in the industry for 15 to 20 years. The desire to achieve this is powerful, but the challenge is that you can’t change culture easily. At the root of this problem is that getting access to information is hard—and if you make a change hard for people, they won’t change their behaviors.
Let’s envision creating a new marketing campaign, ideally this should start by reviewing the success and challenges of the previous one, but how would you find last year’s marketing campaign data? In the enterprise you first have to know which portal or site to go to in order to find the marketing data, then you need to sift through a multitude of marketing reports to find what you’re looking for. If you are lucky and get this far, you may get stuck when discovering that the report you found doesn’t have all the information you need. We don’t make it easy enough for people to get to information. With everyone being busy, they are likely to skip reviewing the past marketing campaign altogether and miss any valuable insights instead to just start from scratch.
Techniques of natural language and conversational interfaces make what was once too hard a trivial exercise. With augmented BI you now have one single place to go, and all you have to do is ask a question. Once again, the system can look across all the reports. It can look across all the data. If that information isn’t already reported on (or even identified), the system can simply compute it. And so, rather than making you go through all those steps, it will simply give you an answer on how last year’s campaign did. From there it might make suggestions: “Would you like to examine how the campaigns did across the different regions?” So not only are you not wasting time searching for where to start, the system can give you answers to questions you weren’t even asking.
The Perfect Collaboration
In an augmented BI environment, we let humans focus on things that our brains are good at: being creative, thinking outside the box, thinking non-linearly. And we let the machines do what they are good at: things like complex mathematics and repetitive tasks. It’s a perfect collaboration between human and machine, and it allows greater collaboration within and between teams within the business. Augmented BI enables greater productivity for everyone in the organization using analytics or BI, not just business analysts and data scientists, but everyone is free to dig deeper and find better answers than a traditional BI workflow would allow.
And the collaboration does not stop there. In an environment where users can ask freeform questions and are encouraged to take decisions on the resulting information, SAP Data Warehouse Cloud is a perfect fit. With its embedded semantic layer that translates data structure into business terminology, along with its ability to unite all the organization’s data across multiple landscapes, SAP Data Warehouse Cloud provides the perfect environment to support trusted data-driven decisions. Together, SAP Data Warehouse Cloud and augmented business intelligence provide a synthesis of data and business unlike anything that has come before.
Learn more about intelligent data warehousing on the SAP Data Warehouse Cloud product page.