How does "big data" strengthen or even replace management?

Marketing executives at a sizable retail company in the United States recently found themselves ignorant of sales reports. One of its major rivals is continuing to expand market share in a range of business areas. Despite a combination of online promotions and optimized sales, her company continued to lose ground in the city.

So she summoned the top brass to do a thorough study of the competition and found that the root of the problem was far beyond their imagination. Opponents invest heavily to improve the ability to collect, consolidate and analyze data from stores and apply them to each sales unit. It also links this information to the database of suppliers, adjusts prices in real time, automates replenishment, and easily mixes products between stores. Through continuous practice, bundling, aggregation and seamless convergence of information in the Organization (from grass-roots stores to the office of the CFO), competitors reinvent themselves as the fastest-responding companies in the industry.

This is the first understanding of "big Data" by the retail executives ' team. Although data has been a symbol of the information age from the outset, the volume has exploded over the past few years. In 15 of the 17 economic sectors in the United States, businesses with more than 1000 employees store an average of 235 gigabytes of data, beyond the Library of Congress. Although a great deal of information comes from financial transactions and customer interaction, information generated from the new equipment and value chain is growing at an alarming rate. Just think about your business: processing machinery built-in sensors are collecting operational data, marketers scanning social media or using smartphone positioning data to understand adolescent consumer quirks. Data exchange may be networking with your supply chain partners, and employees can share best practices on the corporate wiki.

All this new information is of great importance to the business and its leaders. The latest academic research shows that companies that use data and business Analytics to guide decision-making are more productive and have higher net asset yields than companies that do not. This is consistent with our findings that "networked organizations" have a greater advantage, namely, opening up internal information channels and engaging customers and suppliers through Internet Data exchange.

We believe that in the future "big data" can fully become the new assets of enterprises, the formation of an important basis for competitiveness, like a strong brand. If this is justified, businesses need to start thinking hard about whether they can take full advantage of the potential of massive data and try to deal with possible threats. Success requires not only new skills, but also New Horizons-the advent of the "Big Data" era may affect the expansion of management circles and may herald new and even disruptive business models.

In the remainder of this article, we list important ways in which massive amounts of data can change competition: through process transformation, changing enterprise ecosystems, and driving innovation. The following discussion revolves around 5 questions that executives should ask themselves. By rethinking these issues, executives can better understand how big data reverses the assumptions behind their strategy, and the speed and scope of the current change.

1. What happens in a highly transparent world where information is within reach?

As information from all walks of life becomes increasingly accessible, companies that rely on proprietary information as a competitive advantage will be at risk. For example, the real estate industry uses the special channel of transaction information and the asymmetric information such as the buyer's buying and selling behavior to carry on the trade. Both types of information require a lot of energy and financial resources to obtain. But in recent years, online professional providers of real estate information and Analysis Services have begun to bypass real estate brokers, allowing buyers and sellers to communicate directly, forming a second real estate information channel.

Cost and price information from all walks of life is also getting easier. Another major blow to proprietary information is the aggregation of satellite image information by some companies. After processing and analysis, these images include insights into competitors ' production, shipping trends, and other valuable data to understand their expansion plans or business limitations.

A big problem is that the vast amounts of data accumulated by many enterprises are often hidden in various sectors such as research and development, engineering, manufacturing or service operations, thus hampering the timely use of information. In addition, the accumulation of information within the business unit is another issue: for example, many financial institutions suffer from a failure to share data between lines of business, such as financial markets, money management, and loans, or do not understand the internal relationships of various financial markets, or they cannot form a consensus view of their customers.

Some manufacturing companies are trying to break down sectoral barriers by consolidating data from disparate systems, inviting functional divisions that have traditionally been tightly guarded, and even seeking external information from external suppliers and customers to develop products together. For example, in advanced manufacturing industries such as automobiles, suppliers worldwide produce thousands of components. Today, more integrated data platforms enable companies and their supply chain partners to collaborate at the design stage, which is a key factor in determining final manufacturing costs.

2. If you can test all the decisions, how will you change the way you compete?

Massive data brings different types of decision possibilities. Using control experiments, companies can test various hypotheses and analysis results to guide investment decisions and operational changes. In fact, experiments can help managers differentiate causal relationships from simple correlations, thereby reducing the variability of results and improving financial performance and product performance.

Perfect experiments can take many forms. For example, the main online business is a continuous tester. In some cases, they use a fixed part of the Web page to carry out experiments to find out how to increase the user's participation or promote sales. Companies that sell physical goods also use experiments to aid decision-making, but large data can make this approach a higher level. Part of McDonald's stores, for example, installed devices to collect operational data to track customer interaction, store passenger flow and booking patterns. Researchers can model the impact of menu changes, restaurant design, and training on labor productivity and sales.

If such a control experiment is not possible, companies can use "natural" experiments to determine the source of performance changes. For example, a government agency collects data from multiple groups of employees who are engaged in similar work in different locations. Just by putting the information on the public, the backward employees are promoted to improve their performance.

At the same time, leading retail companies are monitoring the customer's store walking and interaction with the merchandise. They experimented with a combination of rich input data and trading records to guide which goods to sell, how and when to adjust prices. Such methods have helped a leading retailer reduce its inventory by 17% per cent, while increasing the proportion of its own-brand products with high profit margins, while maintaining market share.

3. How will your business change if you use massive amounts of data for a wide range of real-time customization?

End-consumer enterprises have long used information to segment customers and carry out targeted marketing. By implementing real-time personalization, large data can help companies make rapid progress in advanced technology. Next-generation retailers will be able to track customer behavior, update their preferences, and build models of possible behavior through Internet click-through streaming. In this way, they can determine the customer's next purchase time, by bundling the selection of goods and provide a money-saving incentive plan, and fine-tuning the transaction, and eventually make the entire sales successfully completed. This real-time positioning marketing can also take advantage of multi-level membership incentive program information to promote the most valuable customers to buy high-margin goods.

Retailing is clearly the ideal industry for data-driven customization, thanks to a surge in the volume and quality of information generated from online shopping, social networking and recent smartphone interactions. But with the increasingly sophisticated analysis tools that divide customers into more accurate micro-groups, other industries can benefit from new data applications.

For example, a life insurance company provides tailored policies for each customer with refined and constantly updated background information on customer risk, wealth change, family asset value, and other input data. Public utility companies that collect and analyze information about customer segments can significantly change the power usage patterns. Finally, according to the nature of the work and performance of the staff to make a more precise distinction between the Human resources department, is going to change the working conditions and the implementation of simultaneously improve employee satisfaction and labor productivity incentive mechanism.

4. How can "big data" strengthen or even replace management?

Massive data expands the algorithm and machine-mediated analysis of the field of operations. For example, in some manufacturing companies, the algorithm analyzes the sensor information in the production line, forming a self-regulating process that reduces waste, avoids costly (and sometimes dangerous) human intervention, and ultimately increases production. In the advanced "digital" oil field, the instrument reads all kinds of data about wellhead condition, piping and mechanical system. This information is analyzed by a group of computers, and the results are entered into the real-time operations center. The latter adjusts the amount of oil to optimize production and minimize downtime. A large oil company reduced its 10%~25% operating costs and staff costs by 5%.

Now, a variety of products from copiers to jet engines can generate data streams that track their usage. The manufacturer is able to analyze input data and may proactively correct software defects or dispatch service representatives to onsite repairs. Some computer hardware vendors are collecting and analyzing this information to anticipate and maintain ahead of time in the event of a failure leading to a customer's operation. This information can also be used to implement product changes, to prevent future problems from occurring or to provide customers with information, and to provide inspiration for the next generation of product development.

Some retail companies have also come to the forefront of the era of "big data": they use "affective analysis" techniques to tap into the vast stream of data generated by consumers using social media, to master the responses of new marketing campaigns, and to adjust strategies in a timely fashion. Sometimes these methods shorten the regular feedback and tuning cycle by several weeks.

Coincidentally。 A global beverage company integrates the daily weather forecast information of its external partners into its requirements and inventory planning processes. By analyzing 3 data points, such as temperature, precipitation and daylight time, the company reduced inventories in a key European market and increased the accuracy of forecasts by about 5%.

The results are improved performance, increased risk management capabilities, and improved insights (which may continue to be hidden and unknown if there is no mass of data). As the price of sensors, communication devices and analytics software continues to decline, more and more companies will join the management revolution.

5. Can you create a new business model based on the data?

Mass data is the birth of a new generation of enterprises using the information business model. They play an intermediary role in the value chain. They find themselves in the form of "discarded data" generated by business transactions to create valuable information. For example, a transport company, in the course of its business, realized that it was collecting massive amounts of information about global product shipments, and started a business unit that specializes in providing ancillary information for business and economic forecasts.

A multinational company has learned so much from its own data analysis of the manufacturing transition process that it has decided to start a company to provide similar services to other businesses. Now, the company collects workshop and supply chain information for a group of manufacturing customers and sells software tools to improve customer performance. At present this service business performance is better than the enterprise's manufacturing business.

In addition, massive data is also greatly transforming the data integration industry. This is an industry that aggregates and analyzes information from a variety of sources to give insights to customers. In the medical industry, for example, a group of new entrants are integrating clinical, payment, public health and behavioural information to form more sophisticated disease information, helping customers control costs and improve treatment options.

With the spread of price information online and offline, entrepreneurs are offering parity services that automatically edit millions of kinds of commodity information. From the point of view of the retail business, this comparison may be a destructive force, but it has created great value for consumers. Studies show that people who use this service can save an average of 10% of the cost.

(Responsible editor: The good of the Legacy)