Successful Big Data Processes projects hinge on data management
By Gary Allemann, MD at Master Data Management
Forrester Research first coined the phrase ‘big data process’ in August 2011 to describe the shift from a fragmented, silo approach to Business Process Management (BPM) to a more holistic approach that stitches all the pieces of a process together, across an enterprise, driving enterprise transformation.
The parallels between their recommendations and those of leading data management experts are no coincidence and the findings, yet again, emphasise the relationships between data and business processes.
It is essential for data and processes to work ‘hand in glove’ to ensure the business functions effectively. If inaccurate data is captured, a business process will fail. For example, if a staff member in a retail store fails to capture or incorrectly captures a delivery address, the resulting non-delivery of the items purchased will create a poor customer experience and additional costs for the retailer.
Business must understand the relationship between data and process to ensure data supports key business processes. Data is an enabler of process and this is illustrated by the fact that any business process uses data, whether the process is to capture, alter or exchange information. Essentially, an organisation’s process supports the ability to function adequately. For example, a process to update client information is in place to ensure the correct information is always available. Should this not be done, it will hinder downstream processes that are dependent on this data.
The emergence of Big Data is partially due to the reality that data volumes are growing by an estimated 55% per annum. Fortunately, it is not necessary for organisations to manage all of their data but rather focus on key elements that have the biggest impact on the organisation’s ability to function and which will deliver the biggest return. An understanding of how data supports key business processes is an excellent start to a focused, business-centric data management strategy.
Both big process and data management must be driven by business stakeholders. IT simply does not have the mandate to achieve the cultural shifts necessary to really drive process changes or to embed data quality as a fundamental principle in the business. Because a business process defines how a business operates and any change within a business process will have an operational impact on the users, business must drive this change, although IT will provide the critical enabling capabilities and technologies.
Forrester discussed using the four C’s to highlight the relationship between data and process:
Customer:
Customer data is hidden inside “oceans of operational data” that must be mined and cleaned in order to improve the customer experience. Customer data can sit in both traditional, internal data sources, such as the CRM or billing system, as well as in Big data sources such as social media platforms.
Business processes and data management need to leverage both Big Data and traditional sources. This will require adaptation of existing processes and policies to address the data management challenges necessary to cope with the complexities of Big Data.
Chaotic:
Chaotic business processes can only be resolved by developing a better understanding of the relationships between process and data. For example, a sales clerk looking to close a sale may not be interested in the delivery process and ability to close the deal may not be impacted by the lack of information supporting this downstream process. However, the failure on his behalf to capture this data will impact the delivery process as delivery will be delayed due to incorrect or insufficient information. This essentially impacts the overall experience of the customer.
Organisations must take a holistic view of the end-to-end process to understand the relationship between data and process.
Context:
The context of information is critical. It is vital to understand the relationship between business events, operational data and operational performance in order to focus data management efforts where they will have the most impact.
Cloud:
The shift to Cloud architecture frequently overlooks the need to maintain data relationships across a fragmented value stream. Organisations need to ensure all data issues are resolved before migrating to a Cloud environment. If this is not done, these issues will continue within the cloud where they may be more difficult to identify and resolve.
The four C’s can be used as a guideline for data governance, and data-centric IT projects such as Master Data Management (MDM) and Enterprise Resource Planning (ERP) implementations. Data quality issues are key indicators of limitations within their current processes. Once these limitations are understood, organisations are able to correct them and further improve how they function as an organisation.
In conclusion, the driver of enterprise transformation hinges on BPM, which can further assist companies, offering key enhancements for organisations. These include increased process efficiency and productivity; continuous process improvement; improved process quality, consistency and compliance; cost reductions, increased customer satisfaction and improved reporting of process performance. Understanding the relationship between data and process is the first step towards improving all Big Data processes.