The Data For Business Intelligence

Published Date: 02 Nov 2017

Firstly, ERP system is said to be an indispensable part of any companies and its advantages is undeniable. The most striking feature from ERP system is the integration and collaboration of all departments. With an ERP system the company’s work flow should be coherent and smooth from each department to another thanks to the use of one shared data format and same software across all applications. We do not need to aggregate information from many systems and transform the data for management to report on the whole organization. Especially, order tracking, inventory tracking, revenue tracking, sales forecasting which are fundamental activities of USAP should be more convenient with the help of ERP system. ERP system plays a critical role in making strategic decision in a big company like USAP. In the other hand, there are also accompanying drawbacks when deploying an ERP system. Firstly, the biggest problem is its relatively expensive cost which consists of software cost and training cost. Gleez.com claims that a full ERP system can run companies millions of dollars, provided that the companies and the systems they're implementing are large enough. In more details, software cost means the expense for planning, designing, implementation and testing. With such a big scale like USAP, the price of ERP system itself is surely not cheap. In addition, ERP system strictly requires employee training to effectively perform all its functions. The success of the system totally relies on how the workers make use of it. Some companies deliberately ignore training or take a short cut to reduce training cost. For that reason, improper training could result in misusing which can cost the company much more. The training cost implies training material, facilities, and human resources. It also takes regular time and this will add up the loss of business. Secondly, customisability is also one of ERP system’s disadvantages. ERP is hardly customisable because the vendor usually does not allow it so. This restricts implementation and adaptation of an ERP system. Nevertheless, business model widely varies by company so it could not easy for a company to choose a suitable ERP system. Above all, it is tough to implement an ERP system on decentralised organisations which becomes more and more widespread. According to Tripsas et al (2001), USAP is of that kind of company because there was no central database of auto parts. Therefore it is not only challenge for retailers but also for ERP system creator. A decentralised ERP system has higher total cost of ownership (TCO).

The implementation of a single vendor's ERP solution in many locations without centralized standardization and control will result in multiple (often widely varied) configurations, or instances, of the ERP application. Each deviated configuration will require unique training and support capabilities. Other areas of non-standardization will also increase ongoing costs, because each unique technical environment will require uniquely qualified support personnel.

Decentralized solutions also multiply the IT landscapes and the attendant system administration staff. Further affecting the decentralized scenario is the work required to periodically consolidate operations and financial information from unique ERP instances.

An added twist occurs if multiple vendors' solutions are implemented at different business units, rather than separate instances of a single ERP product. This causes higher TCO because custom-built integration code, data alignment issues, and variations in vendor-centric business processes require extra time and resources to implement and maintain.

This type of implementation scenario is becoming rarer in today's cost-sensitive IT climate.

Individual business units are feeling increasing pressure to embrace new corporate technical, functional and data architecture standards, especially when an ERP consolidation strategy is driven by new business requirements.

The second topic on which I see the relevance to USAP is Business Intelligence. In this topic, there are four sub-topics that I would like to analyse. They are big and unstructured data, user privacy and data security. As an online provider of aftermarket auto parts, USAP certainly considers Internet data collecting as a part of every strategy. As the result of it, they face the difficulty of big data. Traditional business intelligence vendors are struggling to keep up with the increasing data size. Their tools aren't designed to sift through Big Data. They cannot identify what is meaningful and what should be ignored. They simply present reports on all of it, which quickly becomes a hairy mess of charts, graphs, and grids. They are attempting to update their tools to keep up with Big Data by relying on tools like Hadoop. But with products build for a data world unlike our current reality, these measures are not enough. The volume of data quickly shows BI for what it is: outdated. BI simply cannot put Big Data to work for its customers. Fortunately, there are new tools becoming available which fill this gap. Companies like Tableau are creating solutions which make it easy to explore data and find answers. We classify these types of tools as "Data Discovery" tools--they make it easy (or at least, easier) to find the interesting data within the mountains of available data. The analytics market is bifurcating--splitting into two segments: the old "business intelligence" market and the new "data discovery" market. Gartner Research speaks to this in their latest Business Intelligence Magic Quadrant report.

The data for business intelligence is not only huge in its size, but also unstructured. Therefore, how to structure the unstructured data is also a problem that USAP have face when doing business intelligence. It is difficult to classify unstructured for use in business intelligence because of two main reasons. Firstly, it is not stored on any relational database as well as any format. For example, the comments and opinions about auto parts and company’s customer service crawled from auto forums or auto review websites over the Internet is natural language. That means it has no concrete format or regular availability in any database. However, they are still a very important source for exploring users’ feedback and market trend. Secondly, the data source could be everywhere over the Internet. We are able to retrieve data from multiple sources such as customers’ feedback, social network and RSS feeds.

Regarding big data, security always comes into consideration. Firstly, according to Tankard (2012), it is important to protect sensitive information such as individuals’ detail, intellectual property and trade secret. A centralised data warehouse could be a valuable target for cyber criminals. If users’ information is leaked, the company will lose their trust as well as their reputation. Sony can be taken as an example where personal details of 77 million Play Station users stolen by hackers in 2011. Because of this accidents, the corporation was sued, suffering big loss. Another potential issue is to comply with the law. Organisations need to carefully consider the legal ramifications of where they store and process data to ensure that they remain in compliance with the regulations that they face. To control big data security, Tankard proposed some methods in his report in 2012 such as principle of least privilege, employee monitoring and archiving data that no longer needed. Tankard proposed a security control products such as Vormetric which bring together data encryption and its related policy. For example, USAP can decide who can view the data or even if a person is authorized to physically access the data warehouse, they still cannot read the encrypted data. Employee monitoring is to ensure effective access controls. The system should guarantee that employees’ role changing does not accumulate excessive rights and privileges to violate the rules. For instance, an customer service employee who can access customers’ detail will no longer have that privilege if he or she shifts to human resource department. The aim of archiving and destroy obsolete data is to reduce the data volume, so that the IS system is not overwhelmed, and so the risk of sensitive data being breached is decreased.

Last but not least, business intelligence arouses our privacy concerns. In order to support business intelligence, methods, techniques and algorithms are developed to extract as much knowledge as possible from web data. Hence, analysing users’ behaviours is an essential part of mentioned methods. Nonetheless, this is also disputable because it could be considered as tracking or, in other words, privacy violation. Therefore, USAP should be cautious because it is described as a company that heavily rely on data-driven statistical analysis of website activity (Tripsas et al, 2011)……. (Gartner report)

The third topic that may closely related to USAP business is outsourcing which is a major trend involved the importation of parts into the US by third parties and private label providers from China, Mexico and South America (Tripsas et al., 2011). In 2008, importation constituted about 66% of US auto parts. The advantages of outsourcing includes cost reduction, improving company focus and risk sharing. First of all, the most important consideration in an outsourcing strategy is cost reduction (Farhoomand, 2005). It eludes the need to hire staff because the company can have alternative access to quality service with much lower cost from China, Mexico and South America. They possess natural resource such as which facilitate the mass production. USAP can import cheaper auto parts instead of buying rare earth metals from China or oil from South America to produce more expensive production. Moreover, reducing in-house staff also means limit the size of the company which leads to a decrease in operational cost. In addition to big saving, outsourcing also helps the company focus more on its core competencies. Non-core activities increase with additional workload and the quality of the core activities suffers distraction when your business grows. In that situation, outsourcing to a third party plays an important role by allowing company’s key resources to focus on primary business tasks. For instance, USAP suffered drops in revenue because of the poor reputation in customer service although it is an online provider of aftermarket auto parts. Therefore, getting away from the burden of manufacturing can help USAP utilise its human resources and budget to improve customer service. Thirdly, outsourcing means risk sharing as well as a better risk management. Because the burden is reduced, outsourcing will allow the company to share any associated risks with your outsourcing partners. When companies, both the company and its suppliers, focus on applying their expertise, the risks is minimised. For example - by outsourcing to a competent outsourcing partner you reduce the risk involved in having the same task done in-house by staff that may not be as competent in that field. In spite of possessing advanced knowledge and ideas, USAP cannot be as good at manufacturing as the Chinese partners who are called "factory of the world" with much more experience.

However, outsourcing is not all about the good things, it also has drawbacks that we need to bring to consideration. In 2006, Belcourt indicated four outsourcing issues namely hidden cost, service risks, atrophy risks and operational risks. Approximately 50% of respondents of surveys believe that the actual outsourcing cost is higher than expected. The reasons for this include system incompatibilities. Interestingly, outsourcing did not even save time because one study found that about one-quarter of HR staff time was still spent on which had been outsourced(Rison & Tower, 2005). In terms of service risks, USAP need to understand that the vendor probably enter the market and become a competitor. After a long time outsourcing and acquiring technology, Chinese or Mexican manufacturer would be able to become USAP’s rivals. There is a precedent when Schwinn, a U.S. manufacturer of bicycles, outsourced the manufacture of its bicycle frame to a Taiwanese organization, Giant Manufacturing. A few years later, Giant entered the bicycle market and damaged Schwinn's business (Belcourt, 2006). USAP can deal with this risk by including limit term in the contract to prevent the vendors from using USAP’s technology to compete themselves. More than that, sometimes outsourcing causes atrophy risks, badly influence employee morale. As stated by Babcook in 2004, about 33% of HR professionals are against outsourcing because they think they could lose their jobs, or may be transferred to work for a vendor. Lastly, outsourcing may cause the operational difficulties because it’s hard to manage when the business is not at one place. In 2006, Belcourt suggested that outsourcing employees’ skills could limit the development of the company. Worse, the vendor may even sell the acquired know-how and company secrets to a competitor. For instance, outsourcing IT may reduce the firm's potential for creativity because the vendor cannot know the organization's special needs. There is another example when USAP’s supplier produce low-quality products which can damage USAP’s reputation. Therefore, USAP agents needs to constantly survey the vendors, i.e. quality assurance, to make sure the outsourced auto parts meet the OEM requirements.

Besides three mentioned topics, IS implementation also plays an important role in the development of USAP. The company shifted away from storefront operation to concentrate on online sales. That is the perfect environment to apply information technology. Consequently, the USAP information system was highly recognised for its understanding of what kind of information that support customer making decisions and for its helpful how-to videos (Tripsas et al., 2011). In that report, according to Tripsas, USAP used IT to track vendors to optimise the price. In the future, the company was expected to enormously reduce the long-run costs by participating in many IT projects. At the present, there are two main ways to implement IT projects. Firstly, the waterfall model, according to Thummadi in 2011, "is typically described as a unidirectional, top down, and non-iterative activity sequence". The most striking feature of the model is that each phase must be completed, or almost completed, before the next phase is carried out. That means requirement and design of the project is identified at the early stages. Large amounts of project resources is invested in requirements and design activities and Commitments must be made at an early stage in the process, which makes it difficult to respond to changing customer requirements. Therefore, the process is relatively rigid and any late changes will cost expensively. However, it is especially used for large and complex IT projects because of its undeniable advantages. Because each phase is completed before moving to the next phase, documentation is produced at each phase and that it fits with other engineering process models. The waterfall model reflects the type of process model used in other engineering projects. Apart from waterfall approach, the company has another option which is more flexible – agile method. Compare to the traditional method, the newer has some innovative features. "Agile methods deal well with unstable and volatile requirements by using a number of techniques of which most notable are: low ceremony documents, short iterations, early testing, and customer collaboration" (Ji and Sedano, 2001). Agile is much more disciplined and locally reflective than waterfall. Agile uses fewer activity types which are linked in a more organised and repetitive style. Meanwhile, waterfall was less ordered and also use more of activities type. With agile, the company does not need to run intensively in a particular period, they but need to produce code continuously. This means that companies can adjust agile methods to fit their style with their clients and the prototypes are used. In contrast, the waterfall method makes more use of specifications. These features allow agile methods to deliver business value early with the smallest working function. Moreover, we are able to continually improve it while adding further functionality during the project life (Cohn, 2005). Regarding USAP, their IT projects whose aim is to cut the long-run costs but they increased the short-term expense. As the results of it, they should produce the core deliverables as soon as possible, then improving it by adding supplementary functions. At the same time, because the project is quite complex and need many formal documentations, then its requirement and design should be concrete, features of waterfall model would be applied. Hence, the most suitable solution for USAP would be a combination of two models. In overall, the projects is conducted using waterfall model to manage the progress. However, in my opinion, to deal with the sub-systems, agile method should be the best choice in order to deliver core functions early to reduce the cost.