Click to learn more about author Germán Viera.

In today’s modern workplace, capitalizing data is not a competitive advantage anymore, but a must have skill. Businesses around the world are transforming their data flow processes in order to democratize the information within their own structure. The objective is to provide mid-managers and area managers the ability to access fresh information and analyze it with their own point of view.

This concept, in the tech jargon, is known as Self-Service Business Intelligence. In this new wave, users are no longer tied to Information Technology (IT) areas in order to analyze the organization’s data through fixed reports. IT provides users access to corporate data sources and data marts under their control, while the user decides how to link the information, how to visualize it and how to navigate the different dimensions. This new breed of Business Intelligence (BI) Tools is allowing organizations to be smarter and faster with their data-driven decision processes. The “kill the report” motto is growing among tech savvy managers, for a more interactive way of analyzing data.

But as every new trend, the understanding and ability of an organization of deploying such a new concept faces internal resistances and incompatibilities with current processes and usages.

In this article, we will describe five important pitfalls that organizations and technology consultants should consider before deciding to deploy a Self-Service BI Tool and processes within an organization. The situations described have happened in my own experience of real implementations of Self-Service Business Intelligence projects.

The Business Analytics Process

Before diving into details, I would like to provide some ground about Business Analytics and the information processes in an organization. There are three layers within an organization’s data access. Each layer’s performance impacts directly in the next level. This means that the more mature the organization, the more stable each layer is and the more stable the boundaries are. This maturity characterization will help the reader understand some responsibilities over organizations data and the evolution of Business Intelligence over time.

The Waves of Business Intelligence
Images Source: SlideModel.com

• Layer 1 – Technical BI. This layer is responsible for the collection and storage of data for the organization. Traditionally, data is generated through different systems and stored in databases that will later populate Data Warehouses. This layer is owned by the IT area of the organization. They are responsible for the correct storage, synchronization, and population of the organization’s data marts. This layer decides which are the Business Intelligence core products and which will be the multidimensional solution to be used across the organization. Fifteen years ago, this was the only layer of Business Intelligence. IT needed to interact with different business areas to be able to provide the information they needed for the decision-making processes.

The information at this layer is already processed for fast query, drill down/up and slice and dice. But, in order to query or combine the information, complex queries are needed that only specialized employees are able to carry out. The information is delivered in the form of static reports or pivot tables where business users end up navigating the information. Some example of core products managed by the first layers are Microsoft SQL Server and Analysis Services, Oracle OBI, Microstrategy and IBM Cognos. Whenever new information was required to come into play, the different areas required to align to build a solution.

• Layer 2: The second layer of innovation in BI was the inclusion of analysts through the use of interactive tools. Data Analysts created reports and visualizations to be delivered for end users. This layer allowed IT areas to unload they “reporting” work and focus on infrastructure of data rather the analysis and linking of data. At this point of mature IT areas and Data Analysts are the keepers of data, and work on demand based on end user data requirements.
• Layer 3: The third layer of BI innovation came in to the scene in the last three years; It made BI accessible and consumable for end users, and also enabled anyone and everyone to collect, analyze, visualize and publish data. This third layer is the consumerization of Business Intelligence, providing users the ability to create their own reports, based in the corporate data marts, and with external sources that enrich their analysis. Of course, this layer requires the previous two, and these dependencies influence the deployment of this solutions. Products leading this wave are Microsoft Power BI, Tableau and Qlikview.

The layers described before, interact with data. But, data itself has its own process within organizations. Real Business Intelligence reports cannot consume data from untrusted sources. Users must be sure that the data consumed had the correct treatment and is correct for decision making.

The following diagram describes the data flowing within an organization:

The Data Analytics Process
Images Source: SlideModel.com

The first step of the process consists of Data itself. At this step, IT areas deploy information systems to structure each piece of information in the business process. During the Data Step, systems are isolated within their own responsibility and integration happens through systems interfaces. IT areas and business areas work tightly to be able to systematize all the data flows and adapt the processes of the organization according the data recording needs. Once maturity is reached under this step, the organization is able to work through the systems and provide traceable information about its activities (Financial, Operational, Workforce, etc.)

The second step, called Information, consists of the structuring and delivering of integrated information into the business areas and executives. At this step IT areas work with business analysts and managers to create integrated repositories, reports and visualizations delivering processed and consolidated information to the organization. The integrated view of the business allows better data-driven decision-making processes. This level of maturity is essential in current organizations in order to keep the pace of any competitive market and to comply with regulations.

The Knowledge Step is an evolution of the descriptive nature of Information. IT areas and Analysts combine their knowledge with implementation of Business Intelligence Tools that allow the integrated consolidation and fast query of historical data. With the creation of corporate Data Warehouses and multi-dimensional models, the organization is able to provide descriptive knowledge on “Why” things happened, providing the decision makers the appropriate information to understand the past and prepare for the future. At this stage, the organization’s usage of data becomes more interactive. Management no longer requires flat reports, but they want to analyze data from different dimensions and through different visualization tools in order to identify patterns and derive insights.

The following steps of the data process can be named Insights. At this stage the organization’s maturity starts to show results. User implement Self-Service Business Intelligence Tools that allow them to create their own interactive reports from the organizational data marts. IT areas deploy analytical tools using Machine Learning (ML) that extract from the data marts automatic insights to be employed by users in their analysis. In this way, users combine automated suggestions with their own interactive analysis in order to derive insights that predict future business behaviour. Predictive Analytics empower decision makers into unseen opportunities.

Finally, the last step of the data process consists of what I call the Action Step. At this maturity stage, organizations deploy simulation models that allow users to TEST their decisions against their data marts data, and evaluate different outcomes, with a known level of statistical significance. This level of maturity is the new buzz on Business Analytics and where most of the new tools try to reach companies with Big Data needs.

As a conclusion for the reader, this section tries to explain, the different waves of Business Intelligence within an organization, and the different stages of Data Management maturity.

To read Part 2 of this post, click here.

Photo Credit: GraphicStock.com