As more and more enterprises to adopt big data technology, cloud computing and social media to customer needs to expand their business when they need to master data management (MDM) to follow the trend, extract greater business value from the data, to optimize brand management, develop new channels to market, build their core competitiveness. Effective master data management can not only improve the quality of enterprise data, but also closely integrated and Big Data technologies.
Master data is core business entity
Master data is the business operations plays a key role in the core business entities, they are shared data (such as customer, product, supplier, accounts and corporate sectors related data) between systems, dotted with each company's business systems, within the enterprise can also cross-business, high-value data across systems reuse. It should be noted that not all the master data business data within the enterprise, but it is necessary to share data between various systems is the master data. For example, most of the transaction data, billing data and so is not the primary data. Covering resources, products, events, customers, finance, accounts, employees, partners, etc., used to describe the data is the main core business data entities. Therefore, the identification of the primary data, to grasp the essence of the master data as a core business operations.
In fact, in the absence of the main building data systems, master data is already there but the form of the presence and location of its existence is unreasonable. Master data management can help companies integrate the core, most need to share data from distributed business systems for these different sources of data in different formats centralized cleaning and match, and then as a service to unity and complete, accurate and authoritative master data, distributed system or application needs to use that data across the enterprise.
Enterprise Data Management call the shots most important is to solve the problem of master data consistency between systems, so that enterprises with a unified master data access interface, with focus and content-rich and clean data center, able to provide the business sector consistent, complete information sharing platform that provides a reliable carrier to support business processes and business decisions. Of course, master data management can be used as data standards of an enterprise, strengthen the management of common data entities and model of the business system, to avoid large business concept is inconsistent, reduce application silos, the chance of data silos arise, a fundamental guarantee to achieve a better sharing of data between systems, and enhance the interaction of each system at the enterprise level, so as to improve operational efficiency, improve data quality, the better for enterprise information integration pave the way.
If you talk about the primary data from a product point of view, is how to achieve more innovative products. Our manufacturing company, for example, they would all stages of a product named because of inconsistency from R & D, production and marketing, resulting in inconsistent product information. At this time, from a management point of view, from the point of view will bring products to market quickly a lot of confusion, it is necessary to ensure that the product retains its consistency from the entire information lifecycle. At the same time, also the time to build new products, more targeted, which clearly can fill gaps in the market, which can be derived from new products, which are for the operation of the entire enterprise value basis.
Complement master data with large data
Master Data and Big Data core are data, but their focus is different angle: big data focus on the type of data and the amount of data, master data is more concerned with the quality of the data. Category master data and big data is not static, such as on the microblogging say, is a large category of data, such information is generally not the social class into the master data, but may over time, technology the progress and the cost reduction, soon fall into the category of the main data.
Master Data focuses on the organization of data, and large data provide more raw materials. A precondition for example, now, no matter the product or service sales promotion are emphasized is customer-centric, to promote a policy of insurance companies, will focus on customers and prospects remarks in this policy group, these are big data, but what exactly is specific to the client's remarks belong to the category of master data. Therefore, enterprises should get better application performance and value data, it should be based on primary data center, tendons, through the master data scattered all kinds of big data effectively in series. At the same time, big data is an important complement to the main data, the combination can achieve very good results.
Master data platform is like a octopus, with its main data center, to establish contact with various internal and external systems. Master Data Management not only to integrate high-value data services system, but also to integrate the wider periphery around the main data to the data. ? Large unstructured data often exist in the form, not even within the enterprise, how to combine it Informatica products Greater China chief adviser but Bin believes that companies need a set of products to enable enterprises to take advantage of high efficiency, low-cost computing architecture to achieve: quickly resolve individual identification in large data problems, such as customer identification; in the case of a data set partition feasible, efficient completion of large quantities of data, to avoid duplication of; to better cost-effective way to strengthen the social media analytics, and fragmentation individuals are combined into a complete individual, interactive analysis in order to achieve meaningful.
Now, people do not have the big data like photos and videos to get the master data platform, by integrating the two systems still know where to find this video customer-related information where such a system integration requirements will be more widely . Just past the traditional integrated enterprise core systems, the expansion will gradually get bigger and bigger, more and more rich content, but also in the form of integration between systems also tend to diversify, this is the new trend in recent years. Enhance the ability to integrate and manage enterprise data, data integration will be extended to social media data, network logs, sensor device data, documents, emails, other non-coding format multi-structured or unstructured data, can significantly reduce waste in the garbage time data, big data effectively reduce development and maintenance costs.
Master data management is an ongoing process of building
Although clearly the value of enterprise master data, but often do not know where to start, how to implement. Customers often the hands of a sum of money suddenly asked to be master data, hoping to on-line set of master data management system, but only for one month implementation period of the project. We can say that this is a very unreasonable schedule for a month may have a thorough understanding customer needs, not to mention design platform, and platform integration issues with business systems.
Enterprises using the most important thing is not the technical preparations for the main data plan, but the operational readiness, enterprise master data projects launched in the first project to make it clear to the purpose, goals and vision, give enterprises the business any good. Once you have a business driver, companies need to do a second thing clear understanding of their pain points, because the master data problem is difficult to solve a one-time, there must be a priority, executed in batches. Finally, consider what specific tasks to be done, how to master data management.
One-time master data problem is difficult to solve, resulting in high complexity projects, long duration, spend big problem, pre-construction for businesses perception is not obvious, easy being questioned. Therefore it recommends that companies take a gradual approach to construction, first base, after deepening, first locally, after the cover, so let enterprises to gradually feel the benefits brought by the master data. The first phase of the project can start small scale (such as choosing a business area), a successful business will lead to more business involvement early success will be the driving force pushing up the implementation of the follow-up.
To ensure that companies can implement the project in an orderly manner, the proposed business plan based on the methodology, phased, step by step work. But Bin told reporters that the current Informatica MDM product solutions for master data life cycle is divided into nine stages: visit, discover, model, cleaning, identify, resolve, association, governance and delivery. And this is an iterative process, when a new master data appears or after the master data has not yet reached several stages of quality control requirements, master data management will enter the next cycle. Governance process data throughout the life cycle, reflecting the role of master control for data quality.
More and more Chinese enterprises CIO who has seen the upgrade master data management platform for the enterprise management efficiency, reduce operational risk and IT cost savings and other benefits, and gradually put on enterprise information construction schedule .
Recommended Reading articles
40 + annual salary of big data development [W] tutorial, all here!
Zero-based Big Data Quick Start Tutorial
web front-end development based tutorial
Big Data era need to know six things
Big Data framework hadoop Top 10 Myths
Experience big data development engineer salary 30K summary?