Everywhere we look, data and analytics are at the heart of solving our most common business problems. However, what we refer to as data and analytics no longer fits into a tidy compartment. It’s a broad set of capabilities applied across an even broader, intertwined problem space, where any single problem or decision can easily impact others – not to mention the growth and well-being of an entire organization.
This seems to be reflected in how organizations structure their analytics efforts. In a Mu Sigma survey of data and analytics decision makers at large companies, we asked who is considered to have overall responsibility for analytics at a company.
What we found is that a wide variety of individuals are in charge – the CIO, Chief Analytics Officer, CMO, Chief Data Officer, or the CFO. Sometimes, it was multiple roles at the same time (Figure 1).
Reflective of the myriad roles that are considered in charge of data and analytics, I propose a new moniker: The C#O. And the C#O needs help. Companies have made great strides in getting a grip on their data, analyzing it, and monetizing resulting insights. But today’s analytics climate still feels like the Wild West, as evidenced by the wide variety of owners listed above. Models for governance, prioritization, data and information architectures, talent development, and approaches to problem solving are still ill-formed or even absent. If you are that C#O, the CEO is pointing at you to figure all of this out. And you may not be focused on the right areas.
In the past, the CIO job title stood for “Career is Over” due to the perceived brief tenure of individuals holding that title. Now C#Os also must prove themselves fast to avoid this same fate. As a new C#O, consider the following four pillars on which to build your data and analytics initiative, and hopefully to keep your job well past 18 months.
Establish a Federated Governance Model to Balance Control with Empowerment
Large companies recognize that it’s the Wild West with regard to analytics. And often, when a company wants to gain more control, it looks to centralize. That’s now beginning in the domain of data and analytics. Our survey found that 44 percent of organizations have centralized oversight of analytics, versus 22 percent decentralized; and 45 percent of the overall sample expect to become even more centralized going forward.
In centralized models, a single team owns the data and serves the vast majority of analytics needs across business functions. While this carries the promise of an integrated data infrastructure and economies of scale, it won’t provide the agility or flexibility required to keep analytics relevant in the business.
The risk, of course, is that over-centralization will lead to a repeat of the evolution of other shared services functions – like IT – which in some cases have become considered necessary evils. Organizations that are quick to centralize their analytics resources and governance models run the risk of choking the ROI from their investments.
If you’re the C#O, it’s your job to make sure your governance structure allows people to make the most of your data while maximizing ROI and security. The federated model seeks the advantages of both centralized and decentralized approaches. While business functions enjoy the flexibility to deploy analytics where needed, a governing council ensures broad alignment on data policies and infrastructure, as well as problem-solving tools and techniques. This model is the most difficult to execute due to its matrix nature, but also has the most potential payback and stands to drive the most collaboration in the business.
Become the Axis for Problem Solving in your Organization
As C#O, you’re not just the data czar. Your role is to help your organization scale and continuously learn and improve its ability to solve business problems using data and analytics. In that federated model, there are still critical capability-building activities that must remain centralized under your control.
Consider labeling it a problem-solving center of excellence (COE), with some activities that are arms and legs – flexible data scientist capacity doing work for the lines of business, but also educating on best practices and modeling techniques. The COE also could serve as a lab for incubating and evangelizing new machine learning or business simulation techniques.
But, most importantly, the COE needs to be the steward for consistent problem-solving methodologies that all too often are absent in today’s enterprise. In our survey, 39 percent of respondents said that they don’t approach the majority of analytical problems using a single, consistent methodology. Why do IT and finance departments follow clearly defined playbooks, but not analytics professionals?
Your role as C#O is to establish a healthy habitat in which problem solving can thrive. This means putting in place a standard framework for defining business problems and identifying which questions to ask and which hypotheses to test. It means taking ownership of creating analytical roadmaps, ones that account for the interconnected nature of business problems.
Set up Decision Supply Chains to Drive More Consumption of Insights
In our survey, 47 percent of respondents mentioned that they are currently dealing with data challenges – whether about quality, consistency or availability. And issues related to data were described by more than a third of respondents as the number-one challenge – topping other issues like skill shortages. Too often, C#Os obsess over data at the expense of the decision, telling themselves nothing can be done until the data is perfect.
To be an effective C#O, you can’t afford to accept the premise that data matters more than the decisions. Instead, you must find a way to close the attention chasm between the obsession over data and the ultimate goal of making better decisions faster. You must shift to focusing on decisions, rather than solely looking at the data at hand. Fail to spend equal, if not more, time on the decisions, you may end up out of a job.
One construct that helps reinforce this mindset shift is what we call the decision supply chain (Figure 2). Decision supply chains are similar to their physical brethren. There’s a need to retrieve raw material (data), respond to demand fluctuations, and distribute insights across an organization, all while seeking to continuously improve the flow of the chain.
Putting such a construct in place for every business problem forces a team to obsess early about how the insights and analytics they create will be consumed by business leaders. It’s also a great way to enforce transparency across the links in the chain. For example, rather than a black-box approach to the “Perform Analysis” step, open up the approach so that business partners (your consumers) understand the “how” behind the “why” and “what” of your work. This will add credibility to the chain.
Foster a Culture of Learning and Experimentation
Shifting from a focus on data to a heavier focus on decisions will require cultural introspection. By seeking a learning mindset, where people are encouraged to grow and be flexible, you will see that your teams will benefit more from being able to capitalize on change instead of having to react to it.
It’s also important to focus on an experimental culture, offering the space for teams to develop solutions together and encourage members to fail so they can learn what works and what doesn’t. Among the companies Mu Sigma surveyed, 31 percent agreed that they practice a “fail fast and cheaply” mentality, but could improve on it. As the C#O, you can encourage this mentality by implementing a tight feedback loop. By letting team members fail with the guidance to know what went wrong, you will realize an increase in your team’s creativity, harmony, and efficiency.
In today’s business, it’s all about balancing data analytics and decisions. Many businesses get caught up in the idea that the data will lead to the decision. It’s time for a cultural shift – a shift to understanding that the decision matters just as much, if not more, than the data. As C#O, it’s your duty to take control of business decisions and the data that comes with them. These guidelines will help you lead your business to a more effective problem-solving approach that will result in better business decisions and outcomes, as well as your success in your new role as C#O.
Tom Pohlmann is the head of values and strategy at Mu Sigma, a leading global provider of decision science and big data analytics solutions. In this role, he leads the company’s global brand and communications strategy, manages a portfolio of new client accounts, and oversees the development of programs that continuously align the company’s work with its values and vision. Recognized as an innovator and creative thought leader in the industry, Tom has more than 25 years of experience in strategy and product development, messaging and branding, profits and loss accountability, mergers and acquisition, customer insights and analytics, publishing, and public speaking.
Subscribe to Data Informed for the latest information and news on big data and analytics for the enterprise, plus get instant access to more than 20 eBooks.