10 Business Intelligence Trends: From Analytics Adoption to Explainable AI and Converging BI Platforms

“Business intelligence” has been a staple of an effective corporate entity long before computers appeared on the scene, but leaps in computing power, cloud storage and advances in data processing and visualization have allowed businesses to use their resources vastly more effectively.

The advances in business intelligence (BI) also make it accessible to hundreds or thousands of workers throughout the business, instead of just a few dozen data scientists using specialized skills and software.

Tableau’s 2019 Business Intelligence Trends report outlines 10 of the top advancements in BI technology and how the discipline is being changed by floods of new data and a more distributed mandate to process, understand and make use of it.

1. Analytics Adoption — Analytics adoption means going beyond simply introducing new tools for stakeholders to use in better understanding data relative to their function, but instead ensuring that data insights are understood and being acted upon. Adoption programs in many cases started from the bottom up and are now being absorbed as a fundamental part of BI strategy, empowering many users to make decisions and perform analysis without having to rely solely on IT or data management departments. According to a 2018 IDG Tech Poll cited in the study, 60% of CIOs have plans to spend more on analytics in 2019.

2. Cloud-Based Analytics — Moving databases to the cloud has been a boon for small- and medium-size businesses, reducing their equipment and staffing costs while enabling easy scalability. The increased use of the cloud also adds new wrinkles to the way things are done. One example that the study cites is “data gravity,” refers to the transition of resources toward the source of data as the volume of data grows. What this means in practice is that cloud analytics are becoming more popular for businesses of every size as latency falls and throughput increases for cloud-based data queries and processing. As more processes are shifted toward the cloud, data-sharing across all relevant applications becomes more streamlined and other facets of digital transformation, like IoT integration or unified online product catalogs, become easier to initiate and integrate without having to make extreme investments in on-site infrastructure while still providing unlimited permission access to raw data and analysis tools.

3. Data Privacy — As data collection continues to become a greater part of the public conversation owing to recent Facebook scandals, GDPR regulations and more, privacy and appropriate use and protections begin to take their place at the top of many data manager’s minds alongside metadata, analysis, dissemination and other priorities that give data its true value. One major change is the introduction at many businesses of a code of ethics — long applied to functions like accounting, medicine and law. Many of the ethics principles involved are cross-discipline and apply to data storage, governance and use as well. Many organizations have or will soon institute annual reviews not unlike accounting audits to ensure compliance with ethics guidelines, and managers will also be on the lookout for bias and accurate, factual analysis with an appropriate scope given the available data. Misuse or overzealous application of conclusions drawn from data can be as harmful as having no insights at all. Senior consultant at Teknion Data Solutions Bridget Winds Cogley said “the practice of ethics helps practitioners step back and evaluate a situation from an ethical lens. Above all, data ethics are designed to act as speed bumps in our work so we understand how to face dilemmas both personally and professionally.”

4. Explainable AI — As it stands, many machine learning algorithms and software packages can produce insights but struggle to show how those conclusions were reached, making it difficult for decision makers to implement solutions they may struggle to explain in detail later. Explainable artificial intelligence seeks to solve this dilemma by making AI processes more transparent, allowing users to drill down more deeply into the data to understand how conclusions are drawn.

5. Natural Language Interactions with Data — As data proliferates, keeping up with the growing needs of an organization can become a major challenge for data scientists and IT technicians. With the incredible progress of natural language processing (NLP) over the last few years, however, will allow both experienced and novice users to ask questions about visualizations and other analytics outputs and pivot to related information without having to revert to square one. “Show me the coldest record temperature in New York,” for example, could be followed up with “What about Miami?” without needing to restate the question.

6. Actionable Analytics — Actionable analytics will put data where it is most useful in real time, using software APIs and plug-ins to deliver insights to existing software solutions like Salesforce. A plug-in might pull up the lifetime value of a customer and their buying preferences during a sales call, or a purchase ordering platform might return information on stock levels and expected restocking needs at a predetermined time during the month.

7. Data Collaborations for Social Good — As cloud storage and computing becomes cheaper and more accessible, many enterprises are seeking ways to combine forces and share data to make an impact none could individually achieve. Medical research, like analysis on cancer data, and population trends in housing, employment and other areas are prime targets for these types of initiatives, the study says.

8. Data Management and BI Platforms Converge — As more users are given the training and access to perform their own data discovery and analysis, the role of BI will shift to maintaining high-quality data, metadata and relationships that connect various tools together. This type of data governance will be more streamlined than ever as data warehousing and management platforms are seamlessly integrated into familiar BI tools.

9. & 10. Data Storytelling and Data Democracy — Data storytelling is about more than just helping executives understand casual insights or large data sets; it’s about fostering a culture of conversation around data where many stakeholders are involved in understanding results or data sets that aren’t always cut and dried. Data literacy is becoming an increasingly sought-after skill even in roles that don’t directly perform analysis, while data scientists continue to sharpen their soft skills and bring different perspectives into their purview to understand complex needs of the business and how their visualization and other tools can best be put to use. The trend of democratizing data, the study says, involves data scientists becoming better at applying the data to the business as well as the advanced user (or citizen data scientist) becoming better at analytics so that both can help each other and the business.

You can find Tableau’s complete report and additional insights right here.

Share on Procurious

Discuss this:

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.