Analytics & Insights, Business Intelligence, Self-Service Business Intelligence

4 Tips to Grow Adoption of Self-Service Business Intelligence (BI)

4 Tips to Grow Adoption of Self-Service BI

Businesses today thrive when team members have immediate access to the tools and information they need to make quick, insightful decisions. The key to success has been the influx of self-service offerings designed to empower the business user while enabling IT to shift focus to other strategic initiatives. According to the most recent State of the CIO report, implementing the digital transformation is the most significant focus area for IT leadership.

Of course, self-service doesn’t just happen. It often takes refinement and goes through a number of iterations before businesses realize the potential. Self-service business intelligence (BI) is no exception. The goal has been that business users would be able to get the information they need to make effective business decisions when they need it, and most importantly without the help of IT.

Unfortunately, adoption levels still hover around 20 percent for many organizations. However, according to this CIO article and the results of a recent Dresner Advisory Services report highlighted in this Forbes article, the move towards end user self-service BI ranks among the hottest data utilization trends.

The question is why has the adoption rate remained relatively low? The simple answer is that there’s a perception that self-service BI tools are too difficult to use and often lack the ability to go deep enough for many users. Even as data literacy improves, the reality is that today’s consumer apps offer people capabilities with very simple and easy to use interfaces, and business users naturally want the same level of simplicity. At the same time, information and analytical needs of businesses are a constant moving target – with more questions asked every day, which calls for deeper analysis – all that is difficult for organizations to keep up with.

Getting over these hurdles and increasing the number of self-service users now requires a change in perspective that goes beyond the traditional dashboards, reports, and data visualization that business intelligence has offered for decades.

Addressing search-based analytics

Simply put, for self-service to achieve its true potential, business users need an easy way to ask questions through a dead simple interface. Rather than forcing users to pick from an endless list of column names, dimensions, and metrics, they need to able enter questions just as one would perform an internet search. Imagine a retail sales manager typing “show sales by region for televisions compared to video games for last 6 months” instead of combing through pre-built reports or dragging column names around. Combined with search suggestions and recommendations to guide the interaction, the user experience improves dramatically. Adding a natural language search capability that allows even more free form user input extends the usability even further.

Enabling automated discovery

Automated discovery of insights reduces the business users reliance on analysts—making it easier to answer in-depth questions with the help of machine learning algorithms running behind the scenes. For example, the same sales manager might want to know why sales has changed over time, and they may uncover that certain segments of customers are over- or under-performing. Such as analysis may take hours or days for analyst using traditional methods of data visualization looking for correlations in endless combinations of data columns – or this process can take minutes with embedded machine learning.

This provides business users and analysts alike to take advantage of embedded machine learning to uncover deeper insights in their data like hidden relationships, trends, and segmentation. With automated discovery of insights, users also receive narratives with the data, meaning the interpretation of the visualization is not left to the user to uncover. These discoveries are initiated through a variety of means, from searches, by drilling into visualizations, and simply asking a question.

Extending automated machine learning

Automated machine learning also gives data analysts a way to build and operationalize machine learning models without relying on data scientists. As businesses continue to lean on their data science organization to build predictive models, automated machine learning brings these capabilities closer to the line of business and build up the population of citizen data science in a company.

Tellius, for example, offers automated machine learning so data professionals can build, evaluate, and operationalize predictive models. The platform offers a library of different algorithms to choose from, and models can be operationalized through visualizations in dashboards or embedded into external applications via the Predict API. You can include model training as part of any scheduled automated data workflow as well.

Recognizing (and addressing) individual user demands

When it comes to data literacy, it’s important to acknowledge that within any organization, users exist at all points of the spectrum from pure information consumers to data analysts. For many, consumers make up the biggest population. Although the consumer comfort level continues to grow, the majority within this group are only now at a point where they can utilize dashboards and reports to view performance metrics and learn what is happening in the business. When empowered with access to interactive results, it enables drilling down into metrics, personalizing views, and ultimately using their insights as they collaborate with other team members.

Analysts, on the other hand, prepare data for analysis and come up with the visualizations to be consumed by the business. When ad hoc questions arise, they are called by upon to analyze the data to come with those business answers. In order to effectively meet user demands, it is equally important to understand that every organization defines self-service differently in terms what their users need, as well as how they use technology. As such, defining success up front can play a pivotal role in crafting the self-service journey.

Finally, if you want people to actually use your self-service tools, the user experience is crucial. It needs to be fast and easily engrained in the way that people work today, especially embedded in their business processes and workflows. It’s not just about tools – it’s also about supporting users with guidance and training, and consistently communicating changes and updates to them.

I believe that the future is bright for self-service analytics. The road ahead is filled with great opportunity, and with the combination of the right tools, process, and people, we will see many more organizations make better data-driven decisions with greater agility.

Ready to take the next step? Check out this webinar to learn more about how a new approach to self-service BI could benefit your organization!

share

Leave reply

Read similar Posts

  • Self-Service Business Intelligence

    Is the Opioid Crisis real? Let's investigate at the Gartner Data & Analytics Summit BI Bakeoff 2018

    By Alvin Wong

    Discovering hidden patterns and insights to make impactful decisions is one of the key benefits that we receive by leveraging next-generation augmented analytics platforms. This article talks about one such example - the opioid crisis.

  • Self-Service Business Intelligence

    Why Data Analytics Projects Fail (Part 2)

    By Alvin Wong

    This blog continues where we left of in a recent post on why data analytics projects often fail. Across virtually...

  • Self-Service Business Intelligence

    Machine learning – The missing piece in Business intelligence (BI) puzzle

    By Alvin Wong

    Automation is more useful if it gives you the business insights you need to make smarter decisions as opposed to...