Shulou(Shulou.com)06/03 Report--

Why do companies need a data strategy? Who builds the data strategy? In order to solve these problems, Xiaobian summarized this article, I hope you can harvest more. The following are details of how to uncover these problems.

Definition of Data Policy

Data policies are methods for acquiring, integrating, storing, protecting, managing, monitoring, analyzing, consuming, and manipulating data. A modern, comprehensive data strategy is about more than just data; it's a roadmap that defines people, processes, and technology, and illustrates how data will enable and inspire business strategy.

This includes clarifying the vision for the goal and practical guidance for achieving it, clearly articulating success criteria and key performance indicators that can be used to evaluate and rationalize all subsequent data plans. It does not contain detailed solutions to specific technical problems. As business goals evolve, data strategy is not static and needs to keep pace with technological innovation and the way businesses operate.

Why companies need a data strategy

In view of the background described above, data strategy and employee selection and retention strategy are the same, in order to manage enterprise assets more scientifically. Without a policy, the organization is reactive to data requests from business units. From the foreground business, market, to the finance, supply chain and human resources in the middle and back office, all of them will put forward demands to a department, which may involve some data analysis, master data management, business intelligence, data governance, data quality plan, etc.

As a result, the data department has to understand the connotation of business data requirements every day, try its best to meet the schedule, and maintain and maintain the old tools and systems used internally to ensure their normal operation, which is overwhelmed every day. Once there are problems with data quality, metadata, etc., they will be challenged completely, and even upgraded to the height of ability and trust. Crucially, failure to properly support these initiatives in a timely manner can easily lead to poor business decisions.

Who will develop and drive the data strategy

Don't leave enterprise-wide data strategy solely to the CIO, why? Data is not just an IT asset, it's an enterprise asset, and data strategy is partly an enterprise strategy.

DataPipeline believes that CEOs and boards need to understand the implications and risks of quickly bringing data strategies to life, and set out to build the organizational structure described below to encourage culture and innovation.

Compared with other corporate roles, CEOs should focus on both survival and development, and data is difficult to achieve immediate results, so it is CEO wisdom to balance short-term benefits and long-term development tests. If CEOs cite data when announcing decisions and are familiar with data innovation within the organization, then the data strategy is already half successful, otherwise there is a high probability that everyone else's efforts will be wasted.

CDO is led by CEO and is directly responsible for the detailed path and overall rhythm of data development strategy implementation within the company organization. According to business model, CDO determines compliance requirements, value, speed, process of demand satisfaction, and selection of automated and intelligent technology route. Here, we must pay attention to the speed and quality of meeting business needs. Due to the great challenge of data demand, too many CDOs cannot quickly achieve the results that CEO, board of directors and business departments want to see in a certain period of time and within a certain business scope. Without a good starting point, the work of the chief data officer will lose its rhythm and become trapped in a vicious circle of long-term discussions and seesaws with the business department on the process of data collection and use, making this position a high-risk position. According to DataPipeline observation, many enterprises began to set up CDO positions and try to bring business growth through data. Objectively speaking, this, like other executive positions, is a situation where opportunities and challenges coexist.

The Data Compliance and Standards Committee, led by the CEO and composed of the company's line of business leaders, legal leaders, and chief data officers, details the boundaries of data use, freedom, and data quality standards. Responsible for maintaining the highest frequency of discussion updates (typically once a week) as the business evolves, while using automated tools to synchronize rules to the data system. If business changes are not consistent from the compliance level, they will gradually become bottlenecks that limit data usage. The challenge here is not to let the rules discuss too much and completely, to reach consensus within a certain range as soon as possible, and gradually push the rules to land quickly within some ranges, otherwise the landing of the vision will lose its forward rhythm.

The Data Department is headed by the Chief Data Officer and includes data engineers, analysts, and data scientists. The Data Engineer is responsible for ensuring that business users can self-service use and manage data throughout its lifecycle, using self-developed or commercial tools that meet the challenges of the times. At the same time, it is responsible for automatic and efficient integration of data sources inside and outside the enterprise to quickly meet the data retrieval and consumption requirements of the business. In addition, it ensures that metadata, master data, data kinship and business development are consistent at all times, so that the business can accurately understand data semantics.

They should not only ensure the Load Balancer and stability of the big data platform, but also respond to the calculation and query requirements of the business on the data model at any time. It is also necessary to follow the standards set by the Standards Committee, and ensure the quality of data and early warning as much as possible through manual rules and various algorithms. Last but not least, there needs to be a "pricing system" in response to the needs of business units. Because there is a cost in the development and exploration of data support services, but at present the business department has no perception of this, let alone ROI. In the face of cost, it is easy to filter out real requirements, prioritize them, and sort out ROI in subsequent services. This path is difficult, but it is imperative, otherwise the business value dilemma of the data department will always exist.

Sometimes data departments are led by CIOs without a chief data officer, and there is an art of dividing responsibilities. Each business situation is different, but the CDO's focus is on leading data groups to quickly generate business value from data within the right enterprise. CIOs have broader responsibilities, but their areas of expertise are not at this point.

Business departments should have analysts and scientists who can understand the business in depth and use the tools provided by the data department by themselves. At this time, the threshold for use will be continuously lowered, and the difficulty and cycle of fetching numbers will also be greatly reduced. The skill requirement is generally SQL level. Therefore, business departments need to understand data better, and conceive that data can be applied to their own business development perspective, and then continuously summarize in practice by managing the whole life cycle of data usage. The challenge is how to deliver business value quickly and efficiently from data, decoupling growth from data department efficiencies.

IV. Data fusion strategy in data strategy

Why a data fusion strategy

Companies want to support data applications and data businesses with data in the future, but only if they can quickly extract and use data at any time. There are some barriers in this process, such as technology, organization, culture and so on. If you don't think about these difficulties in advance, you will encounter great resistance in the subsequent efforts towards the goal, and it is even possible to abort the project.

What questions do you need to think about answering when developing a data fusion strategy?

After figuring out why we have a data fusion strategy, we will start with the following:

Who: Who in the organization will participate in data fusion? Are there IT experts with programming knowledge to solve all data fusion tasks? Or do you need to empower business employees to use data fusion tools themselves? Once implemented, what are the potential technical risks and implementation cycles? These questions will have a significant impact on the type of data fusion solution you choose to purchase.

What data might need to be integrated? In DataPipeline's view, we oppose big and all-inclusive and support enterprises to develop very flexible data fusion strategies on demand. Enterprises need to think backwards from a business perspective, sort out from a global perspective what business goals they want to drive through data in the next one to three years or longer, what data these data include, where these data exist, and avoid duplication.

If there are only a few silos of data, the most cost-effective strategy for the enterprise may be to choose a basic data exchange or ETL tool that meets specific needs. However, if you need to integrate many different silos (or different types of data), it is best to use a more comprehensive data fusion platform.

When: When is data fusion? Once an enterprise has a data strategy in place, the next step is to put data fusion strategy at a very high level. Because, if you don't do data fusion, the rest of the work will be difficult, but from the time point of view, you have to do a top-level design, and then proceed step by step.

If you are creating a data warehouse, data fusion may occur prior to analysis. If you are creating a Hadoop or similar technology-based data lake that stores data in its original, unaltered form, you will do some data fusion before running your analytics workload. Many organizations today have data warehouses and data lakes, and choosing which architecture will affect the type of technology required for data fusion.

Where: Where will data fusion take place? Cloud and on-premises are not a contradiction, as many companies now have both on-premises IDC (data center) and on-cloud, or even multi-cloud, architectures. Enterprises need to choose a multi-pronged data fusion tool that can support on-premises deployment, cloud deployment, Docker, and large container platforms. In addition, the tool should be modern and intelligent enough to adapt to the very complex and changing environment of the enterprise.

How: How do you integrate data? The problem is the most complex because it involves tools, culture and processes. As a strategic resource for businesses and organizations, we don't want to train everyone to be a high-level, experienced, trained data engineer. Instead, they should focus on cultivating their data awareness, knowing how to use data, and how to use the value of data.

At the same time, enterprises should adopt flexible, easy-to-use and stable data fusion tools. In the face of even more powerful talents, if there is no matching tool and methodology, it is difficult to be comfortable in the work. Future enterprises need to think more about what tools, what platforms, and what ways of working can enable organizations to unleash data efficiency as quickly as possible.

What considerations need to be made when choosing a data fusion platform

Companies need to combine their own development stage, information construction level and personnel literacy to choose their own solutions.

If there are more R & D personnel, the range of choices will be relatively wide. When faced with open source and commercial data fusion tools, you need to choose according to the needs of the enterprise and the requirements for production. Open source is a good way to start, but for business continuity requirements and some production-level requirements, business tools may be needed.

In addition, for some technical considerations, you can see this article: When building a real-time data integration platform, please add links to the technical considerations.

After reading the above, do you have any further understanding of the data strategy? If you still want to learn more skills or want to know more about related content, welcome to pay attention to the industry information channel, thank you for reading.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.