Business Intelligence Trends for 2020: Advice for Buyers

January 28th, 2020

This article uncovers three market technology trends likely to influence buyer behavior in the Business Intelligence domain. These include augmented analytics, and the automation of processes to simplify product usage, embedded analytics, and big data and IoT. 

Augmented Analytics 

The term Augmented Analytics is used to denote the use of machine learning and natural language processing to automate insights into datasets generated by analytics and BI tools. It’s helpful to clearly understand the problem that this approach is trying to solve. The scale and complexity of corporate data have become more than humans can handle.

 Therefore, machine learning and AI are helping to automate data analysis. BI tools begin by gathering data from multiple different sources. The next step is to remove duplicates and bad data, normalizing the data so that it can be effectively analyzed. The result of this analysis is actionable insight into the business. 

Data analysts and scientists spend a large portion of their time working on data preparation tasks, cleaning large data sets, and preparing them for analysis. Augmented analytics removes some of the dependence on data scientists by using machine learning and artificial intelligence algorithms to automate tasks like data preparation. 

Today, many well-known BI platforms have built-in capabilities to automate data preparation and detect some correlations and anomalies in large data sets. The ultimate goal is for this technology to be capable of independent analysis, and even insight generation and action plan generation. This is slowly becoming a reality, but there is still a long way to go. 

Tip for Buyers 

While the traditional BI focus on reports and dashboards is not going away any time soon, companies should start looking at vendor capabilities in the area of augmented analytics. Most BI vendors are building augmented analytics into their platforms, but it is easy to get swept up in the hype. It’s important to include analysts and data scientists in the product selection or validation process to help verify vendor claims and ask the right questions. 

Begin with the BI investments you have already made. Talk to your vendors about their roadmaps to understand their approach to incorporating augmented analytics and timing. It is also a good idea to look at newer, more innovative vendors to see if switching platforms to benefit from faster access to these capabilities makes sense. 

Embedded Analytics 

Embedded analytics is the integration of BI capabilities directly into a business application or portal. The idea is to provide analytics that are accessible where they can have the most impact. In other words, analytics tools are available directly within whatever business application is being used at a given moment, whether it be Salesforce, an ERP system, or a marketing automation tool. 

Reports, dashboards, and data visualizations are part of the daily workflow of business users. This means that the people who most need data insights—those making decisions—have immediate access to analytics tools without leaving the application they are working in. Analytics are delivered in the context of an application that someone uses every day. 

Tip for Buyers 

Embedded BI is becoming the predominant model in the BI space, and many vendors offer an embedded version of their product. But buying a standalone product and an embedded analytics solution are quite different. Standalone and embedded buying cycles have different buyers and buyer concerns. 

It is of paramount importance that embedded solutions should be flexible and easy to use for non-technical users. After all, the users of these systems may have limited skills in data analytics. 

Furthermore, the ease of embedding into the host application is an important criterion. Ask vendors to explain in detail how the embedding process works, and how exactly the BI capabilities co-exist within the host application. 

It is important to have your technical staff understand how open the product is. Products should provide both a software developer’s kit (SDK) and support for open standards so that the capabilities of the product can be extended if required. 

Big Data and the Internet of Things (IoT) 

Most large companies are trying to figure out how to harness the terabytes of unstructured data, like streaming data, video data, machine data, etc., to improve business decision-making and business outcomes. Business data is no longer only collected from operational and transactional internal systems, but from physical devices like IoT sensors and machines, and human sources like social media, image designers, etc. 

The Bl data warehouse was designed for highly-structured data stored in tables. It cannot comprehend this kind of unstructured data or this volume of data. This led to the rapid ascension of Hadoop and so-called data lakes, or vast repositories of raw data stored in its native format until needed. 

This used to be the province of highly-trained data scientists, but the focus today is on democratizing this data so that business users have access. These business users need the ability to integrate data from multiple sources. 

IoT data is not the same as Big Data. While Big Data is crucial for organizational decision-making, there is usually a lag between data collection and data analysis. IoT data is different. Data is collected in real-time to detect security breaches, correct malfunctioning equipment, and more. IoT data analytics, therefore, depend on managing real-time streaming data. 

Big Data and IoT processing usually require specialized applications that are beyond the purview of standard BI capabilities. Examples of other useful technology include: 

  • Data Lakes: These are unstructured data stores, often created using data from the open-source Hadoop ecosystem 
  • NoSQL databases: Used for storing large volumes of rapidly changing data, or unstructured data that is not a good fit for a relational database
  • Predictive Analytics: Applications that can determine the likelihood of future scenarios
  • In-Memory Database: Very fast data stores for analyzing time-sensitive or streaming data (like IoT data)
  • Streaming Analytics: Tools for analyzing rapidly-changing data, like social media data, in real-time 

Tip for Buyers 

Integration with Big Data and IoT technologies is crucial to creating the single analytics layer in BI tools. BI tools should be able to access data from multiple different technologies such as databases, streaming tools, Hadoop data lakes, etc. and integrate that data so that a common set of data is available for analysis at the point of decision. Ask vendors about how this works in practice. Some key questions: 

  • Does the vendor have pre-built connectors to these key technologies? 
  • Does the vendor offer a high-performance in-memory data engine to help users analyze data faster? 
  • What kind of data ingestion and data preparation tools are available for preparing data for analysis? Are these data preparation tools proprietary and built into the products, or does the vendor partner with data preparation specialists to provide this important capability? 

Conclusion

All the buzz today is around machine learning and augmented analytics. But this is not the only major trend affecting the industry. BI buyers should think carefully about standalone versus embedded BI and understand the tradeoffs. They should also understand that buying criteria in each case are likely different. The role of BI in parsing and understanding big data, particularly that from IoT devices, is still an esoteric topic. This use case is beyond the purview of most standard BI tools and different categories of tools are highly relevant.