Data-driven decision making is an acute topic among organizations across industries and country borders. Behind this enthusiasm is the wealth of data available in today’s business environment – both from publicly available sources and from commercial data brokers – and the many success stories on extracting value from that data which are covered and admired in business publications and industry conferences around the world. Furthermore, with the increasing variety and maturity of Big Data technologies, tackling this vast quantity and diversity of data in search of business value is no longer as daunting for organizations as it once was.
Reflecting this broad interest around data in general, investments in analytics initiatives have been on the rise for a while and this trend continues. However, despite the great-looking graphs, diagrams and dashboards that corporate executives nowadays see beamed on boardroom walls and on the screens of their mobile devices, there is a nagging question in the back of their minds: can I trust this data that I’m seeing here?
Most Organizations Face Issues With Trusting Their Analytics Initiatives
Whether–and to what extent–the information delivered by analytics can be trusted seems like a fair question to ask, and quite a few business people are doing just that. According to a 2016 research report from KPMG, only 34 percent of decision-makers say they have a high level of confidence in analytics concerning their operational data. And for analytics that drive their customer insights this number, at 38 percent, is on a similarly low level (KPMG 2016). The report cites some interesting geographic differences in the level of trust–organizations in the United States have the highest confidence in their analytics initiatives, whereas French organizations have the lowest. But across the board, there are big questions regarding the reliability of currently available information that serves as the basis for decision making, often with far-reaching financial, organizational and reputational consequences.
While substantial investments are being made in the area of analytics, only around a third of the decision-makers are truly relying on the insights delivered by their analytics initiatives. Some natural follow-up questions emerge, including what are the reasons behind this, and what can be done about it?
Poor Quality Of Data Erodes Trust In Analytics
To some extent the low levels of trust in analytics can arise from subjective perceptions regarding a lack of transparency or understanding about how the information is put together, or even from receiving information that goes against the gut feelings of decision-makers. The biggest reason behind the trust gap, however, is poor data quality. This can be actual or perceived, as it may be difficult for companies to know how reliable their data actually is, but the problems it causes are very real.
To illustrate how broad of an issue this is, a study by Experian in 2015 found that 92 percent of organizations find some aspect of data quality challenging for them (Experian 2015). Further highlighting the importance of good data quality as an issue, a 2016 study done by 451 Research indicates that only 40 percent of respondents are confident in their organization’s practices around data quality management, while 95 percent of them agree that both the volume of data and the number of data sources in their organizations will continue to increase (451 Research 2016).
It doesn’t take a complex chain of deduction to state that the above equation does not look very promising. Since it seems obvious that things are not likely to get any easier for organizations and that the role of data quality assurance in their operations will become increasingly important, the attention naturally shifts to exploring what can be done to improve the current state of analytics initiatives.
Supporting Data-Driven Decision Making By Addressing Data Quality Should Be A Strategic Priority
The problem is not that quality of data would somehow be overlooked as an important focus area among organizations, or that its impact on business would be ignored. In fact, 93 percent of companies are actively looking for data quality issues, with 50 percent of them stating that fixing issues before they impact the business is their biggest data quality challenge (Experian 2015). Considering that there is a broad consensus among organizations that quality data is an important topic, and one that is already being addressed at least to some extent by most organizations, why is it that the problems persist, and where does one start looking for solutions?
The answer, as with many complex business topics, can be traced back to strategy. Data is no longer a by-product of the various business processes–if this ever was the case–but a strategic resource, the handling of which defines success in the digital world and requires a strategy of its own. Having a defined and centralized data strategy is the starting point for building successful digital business initiatives. As with any other strategy, this needs to be properly implemented across the organization to create value.
The first step in implementing a data strategy is to establish ownership for data and assign clear responsibilities. After this, organizations must have the foundation in place to tackle not only data quality, but also all other aspects of data and analytics operations in a structured manner.
A central part of this structured approach that should not be overlooked is clearly identifying the business value of having accurate and timely data available to support decision making. Measuring the financial cost of poor-quality data varies from company to company and can be very challenging, but some anecdotal estimates indicate poor data quality can cost companies on average around 23 percent in lost revenue (Experian 2015).
Other opinions on the topic gravitate towards a lost business value of 10 – 49 percent, while there are some indications that this number could even be above 50 percent (451 Research 2016). However, rather than placing too much emphasis on nailing down the exact costs or losses to a Euro, companies should focus on the bigger picture, and at a reasonably specific level, identify and communicate how improving data quality helps their business.
CDO As Enabler Of Data-Driven Business Operations
Currently the responsibility of looking after the various types of data in the majority of organizations still lies distributed across the different functions including IT, finance, marketing, HR and customer services. However, this situation is rapidly changing. While only around 35 percent of organizations have assigned a centralized ownership for their data, 92 percent of CIOs would like to see the role of CDO (Chief Data Officer) created (Experian 2015).
This sentiment is clearly being put to action in forward-looking organizations, and an increasing number of companies have set up or are in the process of setting up a separate organizational unit that is tasked with the data and analytics operations. And, armed with modern analytics tools and solutions like Liaison’s ALLOY™ Platform, this function certainly has access to the technology they need to support a data-driven approach to business. As the leaders of this new function, CDOs have a crucial role in bringing together the technology, skills and processes to make this into reality, and they will shape the way modern organizations operate and leverage their data assets.
As vocal advocates of a data-inspired future, we at Liaison Technologies are working with CDOs in various countries and industries by helping them build their organization’s data strategy, define and develop the capabilities they need, and design and run the solutions that power their data operations. If you wish to find out more about our approach, you can contact our data-experts here.