Big data analysis has become popular and been used in many sectors and continuously expanded to more fields. Such analysis aims to establish correlations between factors that are selected by the analysts. Big data analysis has many advantages as well some weakness.
I would like to talk about the advantages of big data analysis. Such analysis can generate higher profits for companies that supermarkets use big data to study consumers’ habits and locate their products for increasing sales. In addition, insurance companies and banks can use their clients’ data to study how risky their consumers are and charge them with correct fees. The IT companies use the data to collect the information of their users and use their analysis for advertisement and posting most relevant information. Therefore, we can see that the use of big data mainly focuses on studying consumers’ habits. Such analysis method is good at identifying their consumers’ characteristics and groups among the population. This could save costs for advertisement and increase the profits, as it can allow companies to amend their marketing strategies to best match their consumers’ preferences and post advertise directly to their potential consumers who have similar characteristics as their current consumers.
However, such analysis method has weakness and is possible to deliver accurate results. Firstly, it is impossible to model any social phenomenon perfectly accurately, since the real world is extremely complicated and we are highly possible to omit some important factors in our models. Secondly, as the sample is not randomly selected (only current users and consumers are analysed), it is more likely to find out the factors that possibly determine why they become the companies’ users or consumers, than to find out exactly what factors makes people turn away from the companies. For example, there is a shop near an army base. After a while, the shop finds the majority of its consumers are soldiers and only a small proportion of its consumers are local residents. If the shop does not have information about the population of local residents, it cannot conclude being a soldier makes this individual more likely to shop in this store, as it may be just because the size of the army population is much greater than the size of the local resident population. Moreover, because companies lack information of the entire population, especially those who are not their clients, they can fail to find out more potential consumers, especially when the companies are trying to expand their businesses. In addition, individuals’ behaviours can change over time, the market research done from last period is not necessarily useful for the future marketing strategies. Exogenous changes are countless and all can contribute to changing consumer behaviour, this adds more difficulties for companies to spot the changing in their consumers’ behaviour. Furthermore, databases can have many different characteristics, including variable types, sample size, and etc.; different analysis approaches have to be applied for analysing databases with different characteristics. Sometimes, when a small change occurs in the database, there has to be a complete change in the analysis model. There is not any model that can work forever. Frequent updating models can be extremely costly; however, without updating models according to the changes in the real world, models can be ineffective to deliver desirable results.
Overall, I think that data analysis is more effective to maintain their existing clients and increase the revenues gained by these clients than to find potential clients whose data are not available. Therefore, in a monopoly market, big data analysis can be used to its best capability as the data for all clients in the market are available. In addition, the technical problems occurring in modelling can affect the reliability of such big data analysis and it requires frequent updates in modelling, which can be costly. This also gives large companies more power than small companies, as large companies can be cost effective in using big data analysis. To conclude, big data analysis can be used by large companies to gain more competitiveness to compete with small companies.
No comments:
Post a Comment