“Without big data, you are blind and deaf and in the middle of a freeway” – (Geoffrey Moore)
Many organisations have huge volumes of data. Whether structured or unstructured, the ultimate goal for these organisations is to bring their data sources together in order to organise it, analyse it and use it to better predict business outcomes. Often, the answer is assumed to be “let’s build a data warehouse” or “let’s get a cutting-edge analytics tool”. Unfortunately, without a high degree of data maturity, these solutions cannot make any material impact. Quite simply, if your input data is inaccurate and outdated, your data warehouse and analytics outputs will be of little value.
When embarking on the big data journey, organisations must address the pillars of data maturity, namely: data strategy, data governance, master data management, analytics capabilities, and internal staff capabilities. An organisation that aims to use predictive models for decision making must carefully consider these factors and mature them in tandem.
Start by defining a clear and viable data strategy to inform all data-related activities. The data strategy must define how the organisation will use data to meet its business goals and should also include the organisational changes required to meet those objectives. For example: An organisation that aims to use existing sales data to predict future sales per region, must store an accurate location for every sales record.
Data governance includes the information management structures and practices that will support your data strategy. One of the key data governance practices is to introduce data stewards who should be involved in all relevant system implementation projects to ensure data integrity and data quality standards are adhered to. The data steward for customer data, as an example, would ensure that all systems access customer information from a central customer database instead of creating new, localised customer records. Large organisations may also appoint a Chief Data Officer (CDO) to drive the data strategy, and to be responsible for embedding enterprise-wide data governance.
Master data management
The term ‘master data’ describes key organisational entities and their related attributes. In the case of a customer entity, the customer’s contact details or credit score would be examples of related attributes that tell us more about the customer. An organisation may have this information stored and managed in different systems, and at an enterprise-wide level it may not be clear which of those systems hold the most up-to-date information per entity. Often there are only a few people in the organisation who know which system has the most accurate information, while the rest of the company is unaware and therefore likely making decisions based on inaccurate, un-managed master data. Master data management entails aligning processes, systems and people to create a single source of truth, such as ensuring all systems read customer data from the central customer database, where only a certain department is allowed to update that database. Master data management essentially ensures uniformity, accuracy and reliability of the enterprise’s shared master data.
Internal staff capabilities
Internal staff capabilities must be matured at all levels of the organisation. This includes the collection, capture, storage and ownership of data., There may be resistance if cutting-edge analytics tools are introduced to decision makers that do not know how to use them. Training should not be overlooked. The organisation should also run awareness campaigns and other change initiatives to socialise the data strategy, introduce the appointed data stewards, and ensure that new data quality and integrity measures are understood and adhered to. Data management can gradually become a KPI in performance reviews to ensure that consistency and quality of data models is maintained over time.
An organisation that has a clear, viable data strategy, has matured its data governance and master data management practices, and has business processes defined by knowledgeable, data-aware staff members, can have confidence in the fact that its data is reliable. When an organisation has reliable data, it can use analytical tools for descriptive (what happened), diagnostic (why did it happen) and predictive models (what will happen in future) with ease.
Need help improving your organisation’s data maturity level? Get in touch via our contact page here and we’ll recommend a consulting option that’ll best suit your specific needs.