There’s a reason why big data analytics are so successful at some companies, yet fall flat at others. As MIT’s Michael Shrage put it in a recent Harvard Business Review article, it all depends on how deeply the data and tools are employed in the business. “Companies with mediocre to moderate outcomes use big data and analytics for decision support,” he says. “Successful ROA—Return on Analytics—firms use them to effect and support behavior change.”
In other words, analytics really need to drill down deep into the psyche of organizations to make a difference. The more big data analytics get baked into business processes and outcomes, the more likely they are to deliver transformative results to the organization. As he puts it, “better data-driven analyses aren’t simply ‘plugged-in’ to existing processes and reviews, they’re used to invent and encourage different kinds of conversations and interactions.”
You may have heard some of these success stories in recent years – the casino and resort company that tracks customer engagements in real-time and extends targeted offers that will enrich their stay; the logistics company that knows where its trucks are, and can reroute them to speed up delivery and save fuel; the utility that can regulate customers’ energy consumption at critical moments to avoid brownouts.
Shrage’s observations come from interviews and discussions with hundreds of organizations in recent years. His conclusions point to the need to develop an “analytical culture” – in which the behaviors, practices, rituals and shared vision of the organization are based on data versus guesswork. This is not to say gut feel and passion don’t have a place in successful ventures – because they do. But having the data to back up passionate leadership is a powerful combination in today’s business climate.
Most executives instinctively understand the advantages big data can bring to their operations, especially with predictive analytics and customer analytics. The ability to employ analytics means better understanding customers and markets, as well as spotting trends as they are starting to happen, or have yet to happen. Performance analytics, predictive analytics, and prescriptive analytics all are available to decision makers.
Here are some considerations for “baking” data analytics deeper into the business:
Identify the business behaviors or processes to be changed by analytics. In his article, Shrage quotes a financial services CIO, who points out that standard BI and analytical tools often don’t go deeply enough into an organization’s psyche: “Improving compliance and financial reporting is the low-hanging fruit. But that just means we’re using analytics to do what we are already doing better.” The key is to get the business to open up and talk about what they would like to see changed as a result of analytics.
Focus on increasing analytic skills – for everyone. While many organizations go out searching for individual that can fill data scientist roles (or something similar), there’s likely an abundance of talent and insightfulness that can be brought out from current staff, both inside and outside of IT. Business users, for example, can be trained to work with the latest front-end tools that bring data forward into compelling visualizations. IT and data professionals can sharpen their skills with emerging tools and platforms such as Hadoop and MapReduce, as well as working with analytical languages such as R.
Shrage cites one company that recognized that a great deal of education and training was required before it could re-orient its analytics capabilities around “most profitable customers” and “most profitable products.” Even clients and partners required some level of training. The bottom line: “The company realized that these analytics shouldn’t simply be used to support existing sales and services practices but treated as an opportunity to facilitate a new kind of facilitative and consultative sales and support organization.”
Automate, and what you can’t automate, make as friendly and accessible as possible. Automated decision management can improve the quality of analytics and the analytics experience for decision makers. That’s because automating low-level decisions – such as whether to grant a credit line increase or extend a special offer to a customer – removes these more mundane tasks from decision makers’ plates. As a result, they are freed up to concentrate on higher-level, more strategic decisions. For those decisions that can’t be automated, information should be as easily accessible as possible to all levels of decision makers – through mobile apps, dashboards, and self-service portals.