Tesco Stores has obtained a unified, guaranteed IP VPN service for its retail chain stores throughout the entire Central and Eastern European region.CLICK HERE
Technical revolution has showered us with information. The amount of data collected in the databases of one corporation is bigger than all information that humanity had produced until the onset of the 21st century. Managing sets of data this large is becoming increasingly more difficult. That is why, for a few dozens of years, data analysis experts have been looking for a way to use this stream of information.
The appearance of new data analysis solutions speeds up information processing, drawing conclusions and making business decisions by increments, at the same time being a source of two types of innovation: the incremental (complementary to the basic offer of the company) and the breakthrough (leading to company or even market restructuring) type.
An example for incremental innovations is the Kayak travel search engine that uses Big Data to forecast air ticket prices for the subsequent seven days on the basis of historical changes (source: Thomas H. Davenport, Analytics 3.0). This is done by algorithms and mathematical models developed by the company. Users of the portal can make better deals, seeing information about whether the price of their ticket is going to rise or drop, including a recommendation to buy/wait and a percentage reliability of the prognosis.
Companies such as Netflix, and the video rental sector or video-on-demand offers as a whole, are excellent examples of breakthrough innovations based on advanced business analytics. The industry has been revolutionized by Netflix at least twice. The first revolution took place when, by using business intelligence systems, the company introduced a unique offer personalization system as well as a flexible pricing policy that favored incidental customers, who are cheaper to serve. The second one came with implementing Big Data to analyze almost all preferences of the recipients of their services, starting from the genre and character of the movie and ending with... colors and layout of the movie cover or poster.
Big Data's revolutionary potential is so big that it may completely break today's business models within a few years' time. This is why managers need to understand this trend and its influence on the structure and business model of their company and its environment.
State of the art analytical solutions open up new opportunities for every company. However, in order to benefit from them, first you need to deploy new systems properly. How is it done?Read more
Below are 5 fundamental business rules that define the essence of Big Data.
The term “Big Data” simply means that the new analytical systems process huge amounts of data. However, in practice it is not the terabytes of data that decide about its revolutionary advantages. What distinguishes the Big Data platform from other, even the largest “data warehouses” is the ability to process data in various formats, even those completely unstructured – such as the customers' activity in social networks or their buying habits.
Basically, Big Data is a set of trends. Having such a set at hand, companies can analyze a huge amount of data in real time, which enables them to offer real value to their clients, individual or business alike. Thanks to that companies can analyze microtrends in the sale of their products or services and adjust the offer accordingly, maximizing their own revenues and profits.
One of the most common myths about modern analytics is that only big and rich companies can benefit from this type of solutions. Nothing could me more wrong. Analyzing 1 terabyte of data costs a few thousand Polish zlotys, while buying a server to manage this amount of data is an investment of a few tens of thousands of Polish zlotys; however the cloud technology may be of help here, by leasing computing power only for the duration of analysis, without the need to buy servers. Companies that cannot afford to spend the above-mentioned sums can use analyses performed as a service or hire consultants who will complete analytics projects using their own tools.
The mere fact that a company knows more about its surroundings is of no importance. What matters are the business results generated by the collected information. Just like at school, in business it is not the one who knows more that wins, but the one who can extract and use the collected information better.
The Marriott chain is a great example of this phenomenon (source: Thomas H. Davenport, Analytics 3.0). The main source of revenue for the company is selling as many available rooms as possible by offering them at optimal prices. Optimal, meaning ones that will bring profit to the hotel, at the same time being attractive to the customers. To reach this goal, the chain has integrated its booking system with an analytics solution that performs online analyses of service purchase prices in the whole world. That is how they know how much in advance customers book their accommodation, which criteria they look at as well as what is the price they are ready to pay for a room in a given period. This knowledge is used by a global team of experts to optimize revenues generated by sales. It also allows to distribute all rooms available in a given location (hotel, town, region) among people searching for accommodation.
More widespread use of tools and methods for processing big data sets may bring a total change not only to the information management methodology, but also to the way companies are organized and administered. Effective use of Big Data tools in a company requires remodeling its internal structure and setting up a special cell or department that brings together the forces of analysts and managers with the competencies of technology experts.
An important element of the Big Data revolution is also the change in the ways decisions are made, departing from the intuitive model towards methods typically used in scientific institutions that involve making hypotheses and evaluating them by means of a structured process.
It has been at least 20 years since companies have been looking for ways of effective data analysis. In the meantime, some of them have already come to hire entire teams of data analysis experts. Unfortunately, competencies required for effective data analysis with the use of Big Data go far beyond the skills of a typical analyst.
An ideal candidate for a data scientist would be an expert on numbers and statistical as well as computational methods, who understands the laws of big numbers. Additionally he/she should be inquisitive and ready to perform research or even question the status quo. These competencies should be matched by business expertise (e.g. experience in designing marketing campaigns), so that the person could not only skillfully formulate a research problem, but also convince other representatives of the company to the proposed concepts.
Even though data analysis tools have been used in business for at least 40 years, it was only when business intelligence systems and, above all, Big Data appeared that the information processing solutions became available to a larger group of employees. This is quite a challenge, both from the organizational and the technological points of view.
In the era of huge mainframes, programmers developed strategic decision support systems that were used by corporate clients and governments. Complex solutions offered precise, but time-consuming analyses that required input from many programmers. As the transactional applications became more common and more data started to be available, along with the increasing processing automation, data analysis systems (already called business intelligence) came to be an integral part of company operation, becoming one of the basic tools of managers' work. Such systems enabled the processing of structured data, e.g. those coming from integrated enterprise management systems.
As the Internet Economy and social media developed, a new challenge, related to the need to use unstructured data, has emerged. The answer to this problem and, so far, the last phase of the analytics system evolution is a group of solutions collectively called Big Data. Such systems are supposed to analyze varied data sets in a flash, picking up unexpected relationships that would be hard to detect otherwise.