A Survey of Visualizing Business Data

To develop a classification, we looked for predominant and recurring themes in the visualisation and visual analysis literature. Firstly, we selected papers that focus on the visualisation of business data designed for a practical business application. We then divided the papers into three primary categories. The top-level categories we used to classify the publications are:

See Figure 1 and Table 1 and Table 5 for an overview of the literature in these categories. Section 1.2 presents the inspiration behind this classification.

Figure 1. The top level hierarchical classification of the literature survey. Green classifications represent leaf nodes, while yellow represents umbrella classifications (see Section 1.1). Financial visualisation (red) is closely linked to this field, but has been published as a survey by Ko et al.

Table 1. This table shows our sub-classification based on data source. Divided into primary, secondary, and hybrid, the source shows the origin of data behind the research. Papers labelled in yellow are not summarised in detail due to space limitations but are still cited to be comprehensive. Papers labelled in green are summarised in the survey.

Type	Source	Business Intelligence		Business Ecosystem	Customer Centric
		Internal Intelligence	External Intelligence	Business Ecosystem	Customer Behaviour	Customer Feedback
Primary Data	Intentional, Active, Digital Collection	Otsuka et al.			Yaeli et al. Nagaoka et al.
	Intentional, Active, Research Study Data	Burkhard Sedlmair et al. Kandel et al. Aigner Lafon et al.	Bresciani and Eppler Bertschi Keahey	Merino et al. Basole et al.	Dou et al.	Brodbeck and Girardin
Hybrid Data	WebScrape		Ramesh et al.	Lu et al.	Shi et al. Sijtsma et al.	Chen et al. Ziegler et al. Oelke et al. Wu et al. Hao et al. Saitoh Fayoumi et al. Haleem et al. Saga and Yagi
Secondary Data	A Priori Database	Wright Vliegen et al. Bai et al. Nicholas et al. Roberts et al. Kumar and Belwal Roberts et al.	Ferreira et al.	Wattenberg Wu and Phillips Basole et al. Basole et al. Deligiannidis and Noyes Basole et al. Iyer and Basole Schotter et al. Basole et al.	Woo et al. Hanafizadeh and Mirzazadeh Kameoka et al. Wu et al. Sathiyanarayanan et al.	Kang et al.
	Business Process	Du et al. Broeksema et al. Ghooshchi et al. Bachhofner et al. Lea et al.	Hao et al. Hao et al.	Basole Basole and Bellamy
	Business By-product	Gresh and Kelton Eick Keim et al.	Liu et al	Otjacques et al. Ko et al.	Rodden Nair et al.

1.1.1. Business Intelligence (BI)

"The main task of business intelligence (BI) is providing decision support for business activities based on empirical information".

Papers that fall in this category aim to provide a visual design that improves the understanding of a business' internal or external environment. The emphasis is that the resulting visual system is created for the use of a single business as opposed to a whole economy or ecosystem. In the "Business Intelligence Guidebook", Sherman states that BI turns data into "actionable" information. It is this output that businesses strive for through whatever means are available to them. BI is seen as both a process as well as a saleable product. We identify two subcategories in this section: Internal Intelligence and External Intelligence.

Internal Intelligence (II)

Internal Intelligence involves the knowledge of internal business processes. Papers in this category often aim to improve business process efficiency or gain a better understanding of the internal structure of the company. This perspective is inward facing. For example, Kandel et al. explored the role that visualisation plays in day-to-day business operations. The focus is placed on a company's internal operations.

External Intelligence (EI)

External intelligence examines the business ecosystem from the perspective of a single business. The focus is often placed on the business competitors to aid in competitive development or in identifying business operations out of the businesses control. This perspective is outward facing. For example, Hao et al. used visualisation to explore fraud detection data in the banking industry.

1.1.2. Business Ecosystem (BE)

A Business Ecosystem is defined as:

"An economic community supported by a foundation of interacting organisations and individuals – the organisms of the business world. The economic community produces goods and services of value to customers, who are themselves members of the ecosystem. The member organisms also include suppliers, lead producers, competitors, and other stakeholders".

To further understand the definition, Rothschild stated:

"A capitalist economy can best be comprehended as a living ecosystem. Key phenomena observed in nature – competition, specialisation, co-operation, exploitation, learning, growth, and several others are also central to business life".

This topic encompasses research that focuses on an economic community. The literature here differs from business intelligence as the research aims to understand an economy from an external perspective instead of through the eyes of an individual business. The focus is on business networks and their surrounding environments. For example, Basole et al. presented an overview of the telecommunications industry using an in-house visualisation tool called dotlink360.

1.1.3. Customer Centric (CC)

Customer-centric literature focuses on visualising customer data. Businesses are moving towards a customer focused method of operating. This focus ensures that the customers' interests are seen as the highest priority and therefore benefits the business through customer loyalty and superior product development. There are two sub-categories in this classification based on customer feedback and customer behaviour.

Customer Behaviour (CB)

The Customer Behaviour sub-category examines potential customers and identifies patterns in their behaviour so that their actions may either be predicted or utilised to sell a product or service more effectively. The scope of behaviour is broad. Geospatial information can be used to inform the physical movement of people, online tracking data can be used to optimise website sale processes, and customer segmentation information can be used to estimate the future behaviour of customers. For example, Yaeli et al. visualised the movements of customers through physical retail stores using GPS tracking data. This customer path dependency falls into the customer behaviour category.

Customer Feedback (CF)

Customer Feedback research focuses on customers who have used a product/service and have provided feedback through any medium. Often surveys are used, and sentiment analysis is performed on the data to examine the feedback. In other scenarios, direct customer feedback is used (i.e., interviews). This feedback is highly important to customer-focused businesses because it enables vital insight into the reception of a product or service. For example, Oelke et al. presented a visual analysis of web scraped customer feedback data from multiple online sources.

1.1.4. Business Finance

The topic of financial data visualisation could be included in this survey. However, this topic is covered fully in the EuroVis 2016 STAR paper by Ko et al. We briefly summarise the survey in Section 2. Additionally, Rodriguez and Kaczmarek published a "Visualising Financial Data" textbook. As such, it is not the focus of our survey. In addition, joining both a business and financial visualisation survey would be too large.

In addition to the primary classification, we have created a second-level classification that is outlined in Section 1.3.