Blog

Defining Actionable Intelligence

May 22, 2019

Any business looking to leverage data for better decision-making should think in terms of actionable intelligence.

Actionable intelligence essentially describes the ability to respond to data in a way that furthers operational objectives — easier said than done. Forrester suggests that some 74% of businesses are interested in being “data-driven,” although only 29% of those surveyed had successfully figured out how to connect analytics to action. But most enterprise decision-makers don’t need a survey to tell them that a disconnect exists between data and action.

The missing piece is actionable intelligence. The problem is that data, information, and insights are not the same. Data can be structured or unstructured, and come from a variety of sources, like operations, transactions, or events. That data can then be processed into information — essentially, data that are fully comprehensible to the human eye (in the form of, say, a map or a graph).

At the top of that stack are insights. Sometimes, insights answer questions, like “Where is the truck?” It’s only the actionable insights that go beyond merely answering questions, and suggest actions that should be taken. The truck’s current location is known — but does that indicate a delay? Should it be rerouted? Should the customer be informed? Actionable insights come from information that can be used as part of a strategic plan. It means the right people are seeing the right information, at the moment when it’s most useful for a timely response.

 

Defining Actionable Intelligence

 

What Makes for Actionable Intelligence?

Not all data will produce insights, let alone actionable insights. For data to empower people to respond, it’s necessary that the system for gathering, processing, and delivering data from diverse streams creates the right kind of insights. That means tapping into just the data you need, in real-time, while allowing everyone to see the same picture. That’s what creates truly actionable intelligence, which must have the following characteristics:

  • Completeness: Specificity and comprehensiveness represent two of the most important characteristics of actionable data. In some cases, that means having a full, across-the-map overview of your domain, with the right number and types of data sources channeling information to the central platform. That can also mean having the ability to go deeper with that data, to understand why something has occurred.

  • Relevance: Actionable insights are only useful if they map to strategic objectives. Information can’t drive action if it’s telling you about something outside the scope of your goals or control. If insights aren’t relevant, that means it’s time to go back to data collection and ensure you’re tracking the right factors, changes, and KPIs.

  • Speed: Don’t fall into the trap of sending data to an analytics tool that no one ever sees. If insights aren’t happening quickly, they may become irrelevant to the goal at hand. In many operational settings, this means that information has to arrive in real-time, that is, in a continuous flow without a lag between state change and notification. Some forms of business intelligence can rely on stored or historical data — but that information is more contextual than actionable.

  • Accessibility: Actionable insights must be both communicated clearly and shareable across all relevant users. Accessible communication often means data visualization and automated alerts, since both of these can enable insights more quickly and easily. Maps, graphs, and charts transform opaque data into easy-to-understand intelligence. Alerts (when surfaced to the right people at the right time) help decision-makers automatically prioritize their objectives on the fly.

  • Accuracy: Only accurate intelligence is usable. In some contexts, the risk that a team will act on inaccurate information is not only unacceptable — it’s dangerous. Data systems must be designed with accuracy as a primary objective. Hardware and devices must be chosen, installed, and used correctly. Data streams should be configured so that data is amassed and translated without lag or bias. The user should never have to wonder if the information they are using is correct — doubt slows down and muddies the decision-making process.

  • Contextuality: Without context, it’s easy to be skeptical about a so-called insight. Without a wider overview of the situation, it may simply be unclear whether any particular piece of information warrants action. If a status changes, is it significant? Is it outside the range of typical or safe conditions? Actionable intelligence relies on having the right sense of the bigger picture. That can mean incorporating historical data, configuring automatic alerts according to predetermined limits, or simply ensuring that the range of data allows for a sufficiently broad understanding
  •  

    Features of Actionable Intelligence

     

    Why Big Business Cares About Actionable Intelligence

    For many industries, real-time operational decisions are crucial to success. If a shipment doesn’t arrive on time, if a security threat isn’t addressed, if the scope of a building fire is misjudged, there are major consequences. Making the right decision is essential, and thankfully, in the era of IoT and increased connectivity, the time is right for transformation.

    For today’s competitive marketplace, actionable intelligence isn’t an add-on feature. It has the potential to transform the decision-making process. Actionable intelligence is not a new idea. What is new is that it’s never been easier for businesses to get it. Today, actionable insights go well beyond quarterly reports or IT crunching last year’s figures. Emerging tools can utilize data sources of just about every type to create a real-time flow of information. Managers, team members, field agents — every decision-maker can gain a better overview of their domain and achieve the kind of actionable intelligence that makes a difference.