Analytics-Driven Decisions

Optimize performance by converting data to insight and action.

When it comes to making informed decisions, most learning organizations are a lot like people dying of thirst in the middle of the ocean. There is plenty of data from which to potentially make good decisions all around us; Yet, there are significant hurdles to converting the data into something useful.

Numerous data sources are publicly available today.

“More data is available than ever before” is probably the greatest understatement you will read today. Less than a year ago, Bernard Marr, a Forbes contributor,  identified 33 Brilliant And Free Data Sources for 2016. Thousands of other sources are available to those who seek them out. Drupal’s creator, Dries Buytaert’s claim that “Data is eating the world” has been supported with numerous examples, such as those cited by Phil Fersht.

Enterprise data is all around us.

Beyond publicly-available data, enterprises have access to their own internal data which includes a broad array of sources–systems data, network data, application data, HR data about employees, formal learning data in Learning Management Systems, experience data in Learning Record Stores, communications data (email, blogs and communities/forums), support data, development data, marketing data, customer data, sales data, intranet search data, and on and on and on.

There are many opportunities for learning to optimize performance through data.

While data is proliferating and the field of data science is rapidly evolving, many enterprise learning organizations lag  significantly  behind in how they apply new data techniques to better serve their enterprises and customers. Ask an enterprise learning team about their use of data and metrics and you’ll typically hear some references to four levels for evaluating training. You’ll hear discussion of grades, attendance, rating of content, usage of content, rollup reports of individual, team, divisions, regions and organizational activities, and data related to LMS vendor features.

However, the potential for leveraging data is far, far more exciting than what’s typically implemented within learning organizations. For example, high-performing organizations are leveraging predictive analytics to recommend personalized learning paths.

High-performing learning organizations take an analytics-driven approach.

One high-performing electronics distributor has used sentiment analysis to correlate top performance with email responsiveness and sentiment. According to Yvette Cameron, “top performers sent 50% more email externally, responded 40% faster to clients and sent messages with 55% more positive tone than the bottom tier.”  This analysis could potentially lead to automated email reminders (dark learning), as opposed to a traditional course.

High-performing enterprises are adapting systems and applications to drive desired behaviors, as opposed to just recommending them. For example, if analysis shows a problem with Facebook accesses, adapting the company firewall may be more effective than a course that tries to convince people not to use Facebook during working hours.

High-performing organizations are using predictive analytics to enable support teams to better serve their customers. Other high-performers are optimizing the onboarding process for new hires and  analyzing patterns of incoming skills against long-term performance to improve hiring criteria and processes.

Systems of intelligence applied to enterprise learning will differentiate between the top competitors of the future and those that are crushed and left by the wayside.

Dark Learning is analytics-driven.

Enterprises with a high level of Dark Learning maturity don’t rely entirely on educated guesses or qualitative responses. Dark Learning uses modern approaches to extract real data from real systems for real insight, and uses that to provide valuable recommendations for enablement to the business, as well as to create contextual and personalized learner experiences.