Keys to Data Monetization-Informing Employees-Part 1

This article is in continuation from the last article I wrote on Data Monetization . You can read it here.. We begin with the first step of data monetization by educating employees and involving them where applicable.

Initially, organizations use data and analytics to optimize internal processes, reduce costs, and improve decision making. Most organizations have pursued this approach to data monetization for a decade or more. The goal is to give employees timely, relevant, and accurate data so they can gain greater visibility into the business processes they manage and make better, timelier decisions.

The way to inform employees is to establish a data analytics program.

The way to inform employees is to establish a data analytics program. The purpose of the program is to create a repository of clean, integrated data (that is, a data lake and/or data warehouse) that employees query using reporting and analytics tools. By centralizing data and decentralizing data access and analysis, organizations can align employees with a common understanding of shared data elements while maximizing insights and usage.

There are three ways companies monetize data and analytics internally:

• Historical reporting and analysis
• Advanced analytics
• Custom analytic applications

Historical Reporting and Analysis

Most data analytics programs deliver reports and dashboards that summarize past activity—last year, last month, last week, or yesterday. Many companies have operational dashboards that display up-to-the-minute activity through real-time data collection and streaming technology. Most reports and dashboards are interactive—they let users filter, drill, pivot, sort, and visualize the data in new ways to analyze root causes of trends displayed on the home screen of the dashboard.

KPIs. Historical reporting and analysis enables individuals, teams, and entire organizations to monitor, measure, and manage performance against key performance indicators (KPIs) embedded in interactive dashboards. Users can scan a dashboard, quickly identify problems and opportunities, and immediately

Advanced Analytics

Analytical Models. Today, companies are moving beyond historical reporting and analysis. They are hiring data scientists to mine large volumes of internal and external data to make predictions. The scientists create analytical models that companies can use to automate or optimize many core business processes. For instance, the models can improve customer retention, generate online recommendations, detect fraud, optimize work schedules and routes, and prioritize sales leads and mailing lists.

The use of data science, machine learning, and artificial intelligence increases the value of data exponentially. In the hands of capable data scientists, these techniques and tools enable companies to work proactively to address customer needs and adapt more quickly to shifting patterns in the marketplace. Rather than reacting to events, organizations can use advanced analytics to take actions that optimize future activity.

The use of data science, machine learning, and artificial intelligence increases the value of data exponentially.

Custom Analytic Applications

Increasingly, organizations want to build custom analytic applications that combine analytics and actions into a seamless workflow. The applications use machine learning, mobile technology, and advanced graphical interfaces to help managers and workers make everyday decisions with greater accuracy, effectiveness, and timeliness.

Rather than spray data and metrics at users, a custom analytic application gives users just the data they need when they need it. They use predictive algorithms that mine internal and external data to alert users to potential issues before they become problems, and they recommend actions based on historical patterns. These applications close the proverbial last mile of analytics between insights and action.

These [custom analytic] applications close the proverbial “last mile” of analytics between insights and action.

Case study

A retail company with hundreds of stores built a custom analytic application to help store managers use data to work more efficiently. The retailer wanted managers to spend more time on the store floor interacting with customers and employees rather than glued to a computer screen analyzing data.

Rather than present the manager with an array of metrics, the application presents him or her with a specifically tailored news feed that combines relevant and timely insights with tasks. It uses traditional targets to show managers how their performance compares to plan and other stores, with the ability to drill into detail. It also prompts them to complete a staffing schedule for the following week and shows how today weather will impacts sales.

The custom analytics application uses a predictive model that blends historical purchasing and staffing data with promotions data and external data from weather and events databases to automatically generate a daily staffing schedule. The automated staffing model not only recommends the number of staff hours required each day, but also explains the rationale for the recommendation. Giving users a machine-generated recommendation doesn’t normally spur them to take action; they need a common-sense reason to justify adopting an automated suggestion. Once a user validates the proposed schedule, he or she clicks “Create Schedule”, and the application imports the recommendations into the store’s scheduling system.

Most companies have deployed reporting analysis capabilities, and many are now deploying advanced analytics teams to use data more proactively. However, few have built custom analytic applications that blend analytics and operations in a guided workflow. This will change as data analytic platforms open up their APIs to application developers and companies recognize the value of custom analytic applications.

This article series is a part of my ‘Data Monetization’ report that we published at Eckerson Group in 2017. You can download the full report here.