Data-driven businesses are growing at an average rate of more than 30% annually as per Foresters estimates. Other research shows that data-driven decision-making leads to output and productivity that is 5-6% higher. All this points to the importance of a data-driven culture in an organization. Instilling a data-driven culture requires changing the organizational mindset to one based on data-driven decision-making and the addition of the right tools and skillsets to the organization. 

However, the importance of a shared understanding of data through effective Data Governance and common definitions of key business metrics (a Single Source of Truth), cannot be overstated. In addition, Data Credibility, which is linked to Data Quality, is crucial in establishing a data-driven culture. 

Now that we have established these key requirements for a data-driven culture, let’s understand what they mean.  

Data Governance 

Data governance refers to the people, processes and technologies required to manage data assets in an organization. It defines the authority over the data assets in the organization and outlines how these data assets may be used. A Data Governance framework supports the organization’s data management strategy in the storage, management, security and collection of data. 

While initially the aims of effective data governance were compliance and security, data governance now needs to be flexible enough to allow various actors in the organization to make the best use of data for decision-making. One key technology that helps in making data more accessible to different people in the organization who need it is the Data Catalogue. 

Data Catalogue 

A Data Catalogue, to use a simple analogy, is like a library catalogue which describes the various books available in the library, their multiple editions, their locations and a description of each. A data catalogue essentially harvests metadata to create an inventory of data assets in an organization. This can help in self-service analytics as well as compliance, audit and change management in addition to creating shared glossaries of data for the organization to help with data governance. 

A Single Version of Truth 

A Single Source of Truth for an organization can be understood as one view of data which everyone in the organization agrees on as the real, trusted numbers for a piece of operating data. Consider, for example, the inventory management of a product at a company. The sales team may have inventory counts for its sales plans in an Access database or Excel spreadsheets. The company’s order management system will have its own data of inventory on hand in both dollars and units. The same data may also be synced daily in a Warehouse Management System. The finance system may contain weekly or monthly figures of the inventory on hand in dollars. The company might even have a separate inventory management solution which would contain the inventory data for a specific sales channel such as e-commerce. It would be very difficult for the business to arrive at the single source of truth regarding its inventory data and make inventory management decisions based on that data. 

Thus, considering the wealth of data available today, it is vital for businesses to establish which data to use for making decisions. In our example, the company needs to select which set of inventory data will be used for decisions involving supply chain. Business Intelligence tools which provide dashboards and analytics can help to present the data considered as the single source of truth, thus strengthening the data-driven culture. This is especially true in the currently recovering economy where knowledge about where the business stands at any given moment can help control inventory, expenses and determine which promotions and products are driving bottom-line results and which are not.  

Data Credibility 

With the amount of data being used for business decision-making, it is vital that 100% of that data be trustworthy. Using untrustworthy data for decisions can lead to business losses and can hamper the data-driven mindset that a company is trying to establish. Once people lose trust in one category of data, then tend to question many more categories of data. 

One area in which data credibility is a big challenge is HR Analytics. For example, providing credible data about employee turnover can be an issue. There are multiple definitions for employee turnover and nuances as to whether you should include expats and long-term contractors. In fact, establishing the credibility of data is seen as one of the key challenges facing a person in an analytics role in HR Operations.  

Data credibility has also been described as one of the key building blocks for successful Workforce Analytics. Credibility of data can be improved if the organization has a clear data policy in terms of the sources of data it trusts. Data trustworthiness can also be established by considering where the data has been published and how the data collection is funded. 

To conclude, data-driven decision-making can be a source of increased output and productivity for a business. However, achieving this requires effective data governance, a single source of truth for data values, agreement on metric definitions and data governance policies.