Abstract
Private and public organizations have been using data for decision-making. However, these organizations have been struggling in putting into practice the design data analytics projects. In this sense, this chapter aims to present a proposal of a designing business analytics projects with a practical five steps dashboarding cycle. The first step, Business Questions, deals with the scope of a data analytics project creating problem-based questions. The second step, Data Sources, details which are the data sources to be collected. The third step, Extraction, Transform, and Loading (ETL), sets up data source routines of what, where, and when to collect data. The fourth step, Data Warehouse, creates a data repository where data is stored and treated after ETL process. The fifth step, Data Visualization, designs a web dashboard with interactive features such as tables and graphs. This chapter ends with three practical examples in both public and private organizations.
TopIntroduction
The use of data in business analytics projects (BAPs) has been used since organizations' information could be computerized or digitized. In a brief historical search, it is possible to identify that business analytics (B&A) and Business Intelligence (BI) concepts were described by Devens (1868) in his work at Cyclopaedia of Commercial and Business Anecdotes. This document is the first known document explaining how business can extract ‘intelligence’ from data analysis, giving advantages and benefits in strategic level.
A century later Devens’s work, the IBM computer scientist Luhn (1958) publishes a paper titled “A Business Intelligence System” in the IBM magazine. This publication aimed to show how data is used to enhance the decision-making in organizations. Luhn is one of the oldest practitioners in organizing scientifically how data can be organized and used for strategic level besides the common storage practice.
Few decades later Luhn’s publication, Dresner (1989) conceptualized data use in organizations. Dresner was a Gartner researcher and Digital Equipment Corporation (DEC) professional. In his paper, Dresner points out that BI plays a key role and describes a set of tools, methods and techniques to improve decision-making in modern business models. Later, this BI description purposed by Dresner was sharpened and have been used as an official concept in all publications and studies conducted by Gartner.
Recently, Davenport (2006); Davenport and Harris (2017) are some of the most recent authors in the BI field, in special conducting research among re-engineering processes after innovative needs during the 1990 decade. These studies conducted by Davenport have a focus on what he defined as “competitive analytics”, in summary, organizations using data analysis to allow competitive advantage over competitors.
More recent studies have been published on the use of BAPs in in sectors such as human resources (Margherita, 2021), supply chain (Boehmke et al., 2020), finance (Dincer, et al., 2019), healthcare and public health (Mosavi & Santos, 2020), manufacturing (Park et al., 2019), marketing (Liu & Burns, 2018), small and medium enterprises (Lu et al., 2020), Enterprise Information Systems (Sun et al., 2017) and, public sector (Matheus, Janssen, & Maheshwari, 2018). However, these studies present different models, frameworks according to the specific needs of each sector. After analyzing this literature, this chapter presents the three research questions:
- •
How can we develop an applied cycle for BAPs in organizations?
- •
Is it possible to designing the BAP projects into steps?
- •
What are the most relevant steps and tools used in BAP projects?
To answer these questions, this chapter aims to present a proposal for a designing business analytics projects with a practical five steps dashboarding cycle. In this text we propose to go forward and contribute with literature about business analytics projects (BAPs) and propose an applied framework for organizations. The expected audience for this chapter is diffuse and ranging from operational data analysts to managers in charge of strategic decision-making in organizations.
This chapter is structured in five sections as follows. The first section introduces this chapter and its objectives. The second section has a non-exhaustive literature review is presented regarding data and analytics supporting theoretical and scientific evidences to be included in the BAP proposal, in special the most common expected benefits and challenges. In the third section, the BAP five steps are presented exploring tasks, and recommendations of methods and techniques to create dashboards in practice.
The fourth section provides a benchmarking of tools to assemble web/online dashboards, including a diversity of paid and free options. Has three examples of dashboards using the BAP five steps dashboarding cycle proposed. The five section presents conclusions of this chapter with a summary of limitations and future research discussion, and a potential agenda of topics and gaps for following practical and scientific publications.
Key Terms in this Chapter
Data Sources: Data sources are physical or digital repositories where the information associated with the business is found. If in physical format (e.g., paper), this data should be converted to digital format to facilitate use and sharing. Usually, digital data sources can be around an organization such as management systems, strategic systems, SQL or NOSQL databases, and, spreadsheets.
Business Question: Business questions are driver questions that business teams should elaborate in a collaborative bottom-up approach based on the company's strategic planning. These questions aim to provide a guideline and find proper answers to the organization's department or area indicators.
Open Government Data: Open Government Data (OGD) are public repositories provided by governmental entities. This data is freely available to be collected and used as a source of data. Historically, public sector is one of the biggest source of reliable data.
Data Visualization: This is the process of viewing and interpreting dashboards. It occurs through the translation of indicators (KPIS) and diagnosis of the status of a particular object, process, team, department or indicator analysed.
ETL: ETL is the process of (E) extracting information from its sources, (T) transforming it to a common format, and (L) loading the extracted information into a location/repository.
Dashboard: Dashboards are digital panels containing figures, graphics, tables, maps and other digital features elaborated from the data analytics process. It aims to translate a set of quantitative or qualitative analyses into a view that can be easily interpreted by non-tech positions such as executives, entrepreneurs and people associated to the business.
Data Warehouse: Data Warehouse is the central repository for the analytical information to be processed. Usually, this data is uploaded to this central repository and have real-time, near real-time, or sporadic updates.
Data Analytics: Data analytics is the process of analysing and visualizing data to improve decision making. It aims to contribute to organizations in the assertiveness of decisions based on data historical series, ceasing to be common sense.