Sergey Nivens - Fotolia
The story of data and analytics in business is one of disconnects. Despite a widespread understanding of the potential benefits of being a data-driven organization, significant investments in data and analytics are failing to yield the expected results. In fact, a recent Economist Intelligence Unit (EIU) survey of 448 senior executives from an assortment of industries found that only 2% of respondents had achieved "broad positive results," despite 70% claiming that analytics was already "very" or "extremely" important to them.
If data and analytics are such miracle drugs for organizations, why aren't more companies reaping the benefits of them?
That same EIU survey offers an explanation, asserting that great analysis often isn't enough, and that there is a strong need for better collaboration between business leaders and analytics practitioners. In short, as Nate Crisel, head of the Real World Informatics and Analytics unit at Astellas Pharma US Inc. put it: "Having bridges in place is what separates leaders from laggards." These bridges are what allow the analytics insights to flow freely from the analytics experts to the business leaders. Without them, the high-powered -- and, often, high-priced -- fruit of advanced analytics will rot on the vine, as the saying goes.
Building bridges is essential in a data-driven organization
So, what exactly are these bridges? Well, they're both tangible and intangible.
In a tangible sense, decision-makers need to be able to easily access and understand data to make use of it. Such widespread data accessibility is often referred to as "data democratization."
Building bridges can be time-consuming and expensive, but it is often a less daunting proposition than forging the crucial, intangible connection between the business and data sides of the house and the building of a data-driven organization and culture.
Such a culture is one in which decisions are made based on well-researched, empirical evidence. In a data-driven organization, data is integrated as seamlessly into daily work as possible, and every effort is made to minimize the effect of cognitive biases (e.g., confirmation bias) on decisions.
There are many prerequisites to an organization becoming data driven, but perhaps none is more critical than having a culture that enables and, indeed, encourages objective, data-informed decision-making. In fact, in a recent survey of its members, APQC found that 36% of respondent organizations consider establishing a data-driven culture to be their top data and analytics challenge -- the second most frequently cited challenge, in fact.
As explained in detail in APQC's best practice report, "Change Management for Establishing a Data-Driven Culture," enacting an organization-wide cultural shift is not a simple transition; it's transformational. A change of this magnitude does not happen overnight, nor is it a single, discrete event. Rather, creating a data-driven culture is a long-term undertaking that requires careful planning and preparation; top-to-bottom, widespread commitment; and enduring attention and effort. The following practices will make creating a data-driven culture and, thus, connecting the business to the analytics a much smoother process.
1. Start by using a current-state assessment to understand decision-makers' needs, strategic goals and capabilities. This information will help organizations map the analytics program to their strategic goals. It will also allow you to understand senior leadership's concerns and to present the business case for change to them in the language that resonates with them, in addition to focusing efforts on areas that matter.
At technology company EMC, for example, the change team conducted interviews with employees from a variety of functions and locations to understand the current state of the organization's processes. EMC used a structured approach to highlight key areas to include in the business case, to help change leaders anticipate the collateral effects of the change, to identify potential change champions and subject-matter experts, and to create an organization-wide sense of involvement in and ownership of the change.
2. Ensure a mix of business skills (including "soft" skills) and technical abilities on the change team. This helps ensure that the communication between the business and the analytics people is constructive, meaningful to both sides and targeted to solve specific business problems.
For example, biotechnology company Genentech relies on one-on-one daily interactions with individuals to reinforce its message about the value of analytics. The organization hires talent with solid analytics skills and well-developed interpersonal skills, such as active listening, specifically to lead this kind of coaching and engagement. That combination of skills is particularly useful for defusing skepticism and overcoming people's fear of a heavy reliance on data.
3. Enact the change using a wide variety of engagement tactics. Face-to-face meetings and training sessions will help engage the more "high touch" people, while dashboards and access to raw data will appeal to the quantitative minded. Whatever it takes to engage people will prove worthwhile in ensuring the success of the change effort.
Electronics manufacturer Lenovo, as part of its evolution to data-driven decision making, provided access and support to ensure that the human resources (HR) organization could consume and leverage metrics and insights during conversations about the business. To facilitate this, the analytics program team created a self-service dashboard that allowed decision-makers to drill down into attrition data to draw conclusions about patterns and drivers of job attrition.
4. Make change an iterative, collaborative process. Both business leaders and analytics practitioners need opportunities to provide input and to help guide the change toward the best possible outcome for all parties.
For instance, on software vendor SAS's predictive workforce analytics projects, the IT analytics team and HR partners meet every two to four weeks to talk about progress, including where the project stands, what has been discovered and any interim results. The business analysts are the first reviewers of analytics results, since they have both data and domain expertise. Meanwhile, the data scientist's role is to look at the data and identify trends and opportunities. Based on that analysis, the team presents results to the stakeholders and uses their feedback to refine the project.
5. Use an array of targeted measures of success to monitor and improve the changes on an ongoing basis. Try to select both behavioral measures (for example, action items, utilization, number of service requests or number of employees requesting training) and performance measures (such as prediction accuracy, A/B comparisons, cost-benefit analysis or stakeholder satisfaction) that capture the value of the changes to the business. Regularly communicate those measures to leadership and modify the measures as needed to keep them relevant to the organization's goals.
IBM's social analytics team, for example, evaluates its success based on the number of requests for analytics that drive business insights. SAS take a more ROI-oriented approach, comparing the cost of implementing attrition-reducing HR analytics with the cost of higher employee turnover to demonstrate the net positive impact of investing in analytics.
The gist of it
Organizations recognize data and analytics' immense potential, but are struggling to bridge the gap between analytics experts and business leaders, inhibiting their ability to achieve that potential. Building a data-driven culture will allow these organizations to connect the dots and realize the full benefit of using data and analytics. As with any cultural change, shifting toward a data-driven culture is a people-centric activity. Therefore, sound, end-to-end change management tactics are essential.
About the author:
Michael Sims is a research analyst at APQC, a business research and benchmarking firm based in Houston. Follow him on Twitter @MSIMS_APQC and on LinkedIn.
Introduce data science into the finance department
Understand the role of advanced analytics
Use analytics for supply chain collaboration