The era of big data--an era of creating super competitive enterprises

This is a fast-growing era, with the popularity of the Internet, the number of data exponentially growth, the same type of enterprises are springing up more and more! So how in this fast-growing era, stand out, grasp the pulse of the Times? The answer is: Build your business big data! To improve the survival and competitiveness of enterprises, Big Data is undoubtedly a sword, through data analysis, not only can you tell the enemy, but also can let their own enterprise winning thousands outside, so that enterprises in competition with peers, more competitive a big weapon, with good, and even can crush competitors. Big data in recent years, the rise and development has been a huge role, according to the analysis of enterprises with excellent big data ability, the likelihood of making the right decision is 3 times times higher than the competitor, the decision speed is 5 times times faster than the competitor.

When an online video site prepares to launch home-made dramas, critics scoff at their ability to grasp the tastes of the audience. It's hard to think of anyone who would have thought that the company would guide producers, starring selection and screenwriter content and hit the market by analysing big data on their accumulated years of user-viewing preferences, helping them get millions of new users in a quarter and gaining several times the price increase over the next year or two.

Four challenges to building a big data strategy:

We are ushering in an era of data explosion: The volume of data generated by devices and interactions is growing at an average annual rate of more than 50%, and is expected to reach 44ZB (44 trillion GB) in 2020. Global companies are increasingly focusing on the opportunities or impacts of big data, and Bain's big Data industry survey shows that more than 400 large companies in North America and Europe (with annual turnover of more than $500 million), about 60% of companies actively invest in big data, hoping to bring significant benefits. (See Figure 1, "the number, type and speed of global data is exploding")

According to the survey, companies with good big data capabilities are twice times more likely to have financial performance in the top 25 of the industry than rivals, making the right decision more than 3 times times more likely than their rivals, and making decisions at 5 times times faster than their rivals. It can be seen that big data is of great importance to enterprises and society as a whole.

Regardless of how the data changes, they are "gold" or "garbage", depending on whether the enterprise understands the data assets that it owns (or can obtain), and builds a clear big data strategy that generates value at the strategic, operational, and frontline levels. It doesn't make sense to have data that is not sustainable to generate value.

Based on Bain's big Data industry survey, companies today face a lot of difficulty in using big data. It mainly includes four kinds of challenges, such as strategy, talent, data assets and tools.

strategy: Only about 23% of companies have a clear big data-related strategy, decide and know how to effectively apply big data analytics to enterprise operations, and build organizational capabilities, processes, and incentives to empower data analytics to support decision-making.

talent: Only about 36% of companies have a dedicated data Insight team and have the expertise and business sensitivity to data science.

data assets: Only about 19% of companies have high-quality, consistent, easy-to-access and application-big data.

tools: only 38% of companies are using advanced big data tools such as Hadoop, NoSQL, HPCC, and automated data cleansing algorithms.

Key factors for building big data strategies and capabilities:

How do companies build clear big data strategies and key big data capabilities? Based on big data-related project experiences with global customers, Bain summed up 6 key factors in building big data strategies and capabilities for companies.

key Success Factor 1: Discover unique "data assets":

When you drag a movie video scroll bar, the video website is analyzing the overall audience's preference data and directing the next episode's storyline; When you shop in the store, the retailer is analyzing the overall passenger trajectory data and directing the layout of the store and the shelves of the goods, and when the plant uses the machine every day, Manufacturers are analyzing the overall equipment usage habits and guiding the design of the next generation of products, repair and maintenance of the initiative to improve efficiency ... There are many examples of this.

As a foundation for building big data capabilities, companies should identify, evaluate, and manage and continuously expand their data assets as they do other important assets.

First of all, we should conduct in-depth assessment of the current status of enterprise data assets, identify the current source, type and data preparation of data assets, assess whether the data is complete and directly related to business development;

Secondly, according to the evaluation results and business strategy objectives, it should be clear what data assets and objectives there is a significant gap, to compensate for the difference in priority;

Then, to identify and evaluate all the external and internal data assets that can be further acquired, after taking into account relevant factors such as data quality, importance and relevance, acquisition cost and time requirements, select the best way to acquire data assets, such as self-collection and collation, external sourcing data, and exchange with external partners;

After acquiring new data assets, enterprises also need to establish data governance mechanism, clean and store data properly, ensure data availability and consistency, and clear data authorization and update system.

Key success Factor 2: Identify how data Assets "Create value":

After evaluating the enterprise's data assets, it is necessary to determine how to use it to support and lead the enterprise strategy. In particular, big data can bring five value to the enterprise:

Optimize internal operational processes: for example, a beverage company uses complex algorithms to analyze big social media data, identify brand opinion leaders with influence on important issues, and target them to enhance marketing effectiveness; a chain of retail companies through the analysis of a large number of store sales data, to find the unknown link between products, To enhance the bundle sales.

Optimize existing products and services: For example, an entertainment company uses an electronic park pass to capture visitor activity data in its theme park to optimize the visitor's experience in the park; a vehicle safety information system service provider uses sensors to collect vehicle driving data to improve its product design, production and maintenance processes.

Development of new products and services: For example, an insurance company uses plug-in equipment to collect driving behavior data, by analyzing drivers ' driving habits to provide a corresponding discount to their insurance, to proactively retain the safer driving behavior of customers; an online film leasing provider analyzes the viewing profile data to improve the user's viewing experience. and provide analysis results to the film investors to optimize the film production content.

Establish a new business model: for example, a medical insurance company through the predictive analysis of patient information data, to the vulnerable population to provide preventive care services to improve the profitability of such customers; a financial services company to give free personal financial software to users, in the user's use of analysis of their consumption data, And then push it to the exact ad.

Gain ecosystem control: for example, an enterprise software company intelligently manages and analyzes the operational data of the channel partners, identifies the qualifications and capabilities of the channel, and forecasts and alerts the performance; an e-commerce company data product team based on their e-commerce platform precipitation of a large number of transaction data, Develop a wide range of big data products for sellers on the platform, helping it achieve data-based operations and revenue, and enhance the attractiveness of e-commerce ecosystems to sellers.

Key success Factor 3: Identify priority scenarios:

Big Data can help create these five strategic values for the company's Business units (marketing, sales, and services), and big data can help optimize internal operations for the company's functional departments (Research and Development, supply chain, and human resources). To identify potential big data scenarios for the company's business and functional departments, you can use brainstorming and in-house seminars to list all possible big data applications by benchmarking the industry's big data application practices, assessing the status of data assets, and profiling data and applications that may be further captured in business and functional processes. It is important to note that big data scenarios must be aligned with the real needs of the business and functional sectors, and should not be divorced from reality.

Table: Examples of common big data scenarios in various fields

Once you've identified potential big data applications, you can evaluate and prioritize big data applications with two dimensions of value creation and business maturity to drive the implementation of the relevant big data applications in sequence. For value creation dimensions, it can be used as an evaluation criterion for creating value, such as improving operational efficiency, improving return on investment, etc., and for business maturity dimensions, the availability of required data assets, as well as the requirements for resource inputs and big data capability support (such as funding, talent, and cross-sectoral cooperation) are evaluated.

Figure 2: Priority evaluation and sequencing framework for big Data applications:

Key success Factor 4: Data-analytics-insight-Decision support for the product, normalization:

In order to efficiently apply big data to the daily operations of the enterprise, it is necessary to constantly translate the data analysis capabilities into internal application products and normalize the data analysis work. For data analysis, it can be driven by big Data application strategy planning, Big Data application scenario design, analysis of big data to gain insight, and drive the design, development and application of big Data application products, and finally realize the sustainable operation of data analysis products. For the normalization of analytical work, we need to continuously maintain the data analysis products and monitor the actual use effect, provide data analysis support for the business and functional departments, and answer the questions in the daily use.

Taking a home appliance company as an example, the data model of demand forecasting and user activity is established by means of collecting big data platform which stores hundreds of billions of users ' data. Based on this, the company has developed an application software with precise marketing capabilities for marketing and sales personnel that can assist in the precise marketing of the region, community and individual users, as well as the development of a user-interactive application software for developers to more fully understand user pain points, popular product features, User interests are distributed with active users who can participate in the interaction. These big data products have made great strides in everyday applications, and in the last year of the system's operations, the company has conducted hundreds of accurate marketing campaigns based on data mining and demand forecasts, with sales of 6 billion yuan.

Key success Factor 5: Provide strong assurance and support for big data through organization, talent, and it:

Big Data strategy can not be landed without the support of organization, talent and it ability, and these key elements and capacity building, need to be close to the line of business, but also with the strategy of consistency.

For big Data organization operation mode, because of big data core analysis ability, tool investment and so on each business unit synergy effect prominent, enterprise (especially large enterprises) generally adopt centralized Operation center mode. At the same time, the rights associated with business decisions and applications are delegated to the frontline department to ensure seamless data analysis and business decision-making. Regardless of the design of a big data organization, the core principle is to ensure that big data analytics capabilities are most effective in supporting first-line decisions based on the company's own circumstances and needs.

In addition, big data organizations need a diverse team of key competencies to work together to support the operation of big Data organization architectures. The talent teams they need include the big Data application Business Manager team, the Big Data analytics team, the data asset Management team, the technology development and maintenance team, and the risk management team, among others.

In addition to organization and talent, big data needs to be supported by a strong IT system architecture. Enterprises need to set up a powerful big data analysis platform system, from different data sources to transfer and analyze data, pull-through data base analysis, in order to unify the service departments of big data application scenarios. At the same time, the data platform also needs to have a unified cleaning and storage across data source data to ensure data availability and consistency of the ability. In addition, enterprises can establish or optimize master data management system, provide unified and convenient data online transaction service for big data analysis platform and business big Data application to support enterprise Big data operation.

Key success Factor 6: Eliminate legal and consumer awareness risks through big data privacy and security management:

While big data brings opportunities and value, it also poses a commercial risk, particularly in the case of laws (such as a social networking platform that is sued by the U.S. Trade Commission for violating its privacy policy) and consumer perceptions (such as the fact that many of the photos taken by an internet company are related to the privacy of local residents, Were heavily protested by the latter). To this end, according to the data type and from  to analysis use each link to identify different types, regions of the regulatory and cognitive risk, and to respond in a timely manner.

Take the  link with a high degree of privacy as an example: for personally identifiable data (such as identity number, etc.), because the law provides the highest level of protection, so if no clear use is not recommended to collect, for sensitive data (such as transactions and credit information, etc.), needs to clearly inform the user and obtain their consent , for non-sensitive data (such as product data, etc.) can be collected on demand.

In addition, enterprises should establish a unified international policy and regulation team, through the data process based on global standards to manage data privacy, and on this basis, according to different laws and regulations of the legal data privacy of the local management. At the same time, we can reduce users ' concerns by proactively disclosing customer privacy policies to obtain data usage analysis authorization, providing users with their own privacy information control and deletion rights, or integrating personal privacy data into group Anonymous data for analysis and obtaining third party privacy risk management certification.

In the process of building big data ability, enterprises need the help and support of professional company. Bain's complete Big data methodology empowers companies to build winning big data strategies and capabilities.

Figure 3: Bain provides big data strategy development, capacity building and decision support and analysis outsourcing services:

The rapid development of big data is both a challenge and an opportunity for enterprises. Companies must seize the strategic opportunities brought by big data, develop clear big data strategies, build strong big data decision support systems and capabilities to fully exploit the enormous business value of the big Data era.

Six typical applications for big data:

1. Personalized Marketing

"We are going through an epic transformation from the industrial age to the information age," said Rogers and Piper, author of the one-to-person future, in the book. We also witnessed the decline of the sales of the public salesperson, one-to-one marketing rise. "In fact, behind the marketing revolution, big data was the initiator. This is also one of the most typical applications of big data in today's business. It can be said that data-driven personalized marketing is becoming an important trend that any enterprise cannot avoid.

Along with the information overload and the consumer heterogeneity, on the one hand is the massive data and the massive information causes the user information hunger feeling, the user to the non-related information tolerance and the day is reduced. At the same time, the user interest data is increasing, but the user screening information capacity ratio and the day, consumers appear long tail trend, all this, resulting in personalized into big data application direction.

This personalized technology is focused on and applied, and in turn, the enterprise in the production field from the simple pursuit of cost-optimized large-scale production to customer-specific direction of change. At the same time, personalized recommendations, mobile cross-screen recommendations become a typical application. Behind these applications is the integration of computer science, statistics and marketing.

2, the identification and mining of customer value

In accordance with Kotler's definition of customer lifetime value in 1995, the customer's lifetime value is "The total amount of sales from a customer's life cycle to reduce the net value that the company spends on acquiring the customer and selling and serving the customer." "is the net present value of all future cash flows that the company will receive from the customer.

This means that data-backed customer lifetime value evaluation and analysis will help the company to establish a market segmentation strategy, to identify which type of customer is worth the cost to build customer relationships, and ultimately find their own real target customer base. At the same time, it will help companies to better promote customer relationship management, such as through data mining and analysis, you can know how much of the sales are from existing customers and new customers. Of course, it will also affect the pricing behavior of the company, such as the price can quickly increase the old customer retention ratio and the new customer acquisition ratio, but also reduce profitability. The price increases will increase profitability, but it will also lower the old customer retention ratio and the new customer acquisition ratio, which means that companies need to use data to support decisions and ultimately get an optimal balance.

3, Customer loss warning

In the era of user-asset, the early warning of customer churn is of great significance to the strategic development of the enterprise, whether the loss is the target customer, whether these customer churn represents a strong offensive of the attackers, or a process of natural selection, which type of customer, or meet the conditions of users more easily lost, The users who meet the conditions are not easy to lose, and so on, through different algorithms, you can find the ultimate customer churn and its reasons, and ultimately help enterprises decide whether to retain these users.

4, data-driven precision advertising

It is noteworthy that in the era of big data, marketing theory is undergoing an important evolution, in history, including products, prices, channels, promotions, such as 4P theory, because Kotler innovation into 6P, plus power and public relations. In 1990, Robert Laute, an American marketing expert, put forward 4G theory, which is based on consumer demand, and re-sets four basic elements of a marketing mix, namely, consumer, cost, convenience and communication. But big data, especially from causality to correlation, is becoming more and more timely, namely, relevance, reaction, relationship and return.

The transformation of marketing theory is precisely the advent of the data-driven precision advertising era, which requires advertisers to identify target consumers before placing ads, in order to achieve accurate positioning, after delivery to use a series of data tools for advertising effect monitoring.

5. Business decision-making

As mentioned earlier, many important business decisions of enterprises have become inseparable from data, in many enterprises, with data to speak, take the data of the road has become the necessary guidelines for business operations. Also in the case of Suning, its data department needs to provide multiple services to the business unit. The first is Reporting Services, which provide real-time, rich, and accurate data support to the operations department through ADHOC Technology, helping all operations to speak with data. For example, if you are doing an activity today, you need to take the traffic data out directly. Second, is the engine service, the engine means that it can use the technology of big data to drive the front office business, it is the difference between the Reporting Services is that it has been directly embedded into the enterprise's production and operation activities, the data will directly affect the business of the entire enterprise.

6. Inventory Management and logistics distribution

For many e-commerce companies, or companies, inventory management and logistics distribution is becoming an important competitiveness of enterprises, it is not only directly related to the cost of enterprises, profits, but also directly related to user experience. Thus, through the data analysis and mining, can accurately calculate the different categories of goods of different specifications of the inventory level, while obtaining the time efficiency of logistics delivery, the best user experience and logistics overall distribution benefits balance.

The era of big data--an era of creating super competitive enterprises