Actionable Insights for Improved Business Results

Actionable Insights for Improved Business Results

The data economy is increasingly embraced worldwide in every industry. According to the World Economic Forum (WEF), by 2025 it is estimated that 463 exabytes of data will be created each day worldwide – that is the equivalent of 200 million DVDs of data created per day [WEF, 2019]. Faced with overwhelming amounts of data, organizations across the world are looking at ways to derive insights from data analytics for improved business results. 

But why do data insights matter for business? Fundamentally, insights help companies gain better visibility to make faster decisions for improved business results. Research by McKinsey Consulting found that companies that are insight-driven report above-market growth and EBITDA (earnings before interest, taxes, depreciation, and amortization) increases in the range of 15 to 25 percent [Mckinsey1, 2022].

What exactly is an insight? According to the Cambridge dictionary, an insight is the ability to have a clear, deep, and sometimes sudden understanding of a complicated problem or situation [Cambridge, 2022]. In terms of data analytics, an insight is the unknown element (such as relationship, patterns, categorization, inferences, prediction, averages, correlations, variations, quartiles, outliers, confidence levels and intervals, and more) that will influence the decision if it becomes known. 

Technically, insights can be classified as hind-sight (understanding a past situation), near-sight (interpreting the current situation) and fore-sight (predicting a future situation) [Southekal, 2020]. However, from the data and analytics value realization perspective, there are two main types of insights.

  1. Performance Insights. Performance insights provide visibility on the measurement entity. Examples are the top five products by sales quantity, top three customers by MRR (monthly recurring revenue), and so on. Effective data storytelling techniques hold the key to managing the two options or choices that performance insights offer: (a) to know or (b) to act. 
  2. Actionable Insights. Actionable insights which are based on performance insights provide the visibility that can be actioned. Actionable insights involve three elements (a) decision (b) commitment to consume resources like time, money, labor and so on associated with the decision (c) and business impact and consequences of the decision. Examples of actionable insights are: product A is the best option to invest in, customer X can be given a credit, and so on.

So, what can enterprises do to get value from insights? In data and analytics, the “last mile analytics” is considered the missing piece between the data analytics output and actual business results. In fact, according to McKinsey Consulting, 90% of organizations that are significantly outperforming peers are devoting more than half of their analytics budgets to bridging the last mile of analytics [Mckinsey2, 2018].  

In other words, business enterprises have to focus on actionable insights – or specifically focusing on converting performance insights into actionable Insights. In this regard, below are the three key steps  companies can take to  implement actionable insights.

1. Derive the performance insights based on business objectives, questions, KPIs (Key Performance Indicators), and data. These insights can be derived using a combination of descriptive, predictive and prescriptive analytics techniques or models which provide visibility into the past, current, and future states.

Given that data is a critical component in this step, the data related to the question and KPI should be sourced from the right data source, typically the transactional SoR (System of Record). In this regard, the semantic layer can be leveraged for standardized data definitions and rapid data access to derive faster insights [Southekal, 2022].

To ensure that these performance insights are reliable, it’s important to factor in different stakeholder perspectives, time frames and location, and avoid the framing bias. Framing bias refers to the manner in which the question is framed and can be addressed by reframing the problem in at least three different ways. Basically, having the right data means mapping the data source subjects and understanding the data attributes (i.e., features) in order to align data source content to the question / answer being sought. The relationship between business objectives, questions, KPIs (Key Performance Indicators), semantic data, and models to derive insights is as shown.

Insights Value Chain

Figure 1: Insight Value Chain

2. Once the performance insights are derived, we formulate the decision problem. A typical decision problem has four key elements: objectives, alternatives, outcomes, and payoff.

    1. The objectives, which are based on performance insights, are the things the business plans to achieve from the decision.
    2. The alternatives are potential actions or strategies considered based on different performance criteria such as profit margin, cost, time, quality, service, and more. It is recommended to keep the number of alternatives to three and the Pugh Matrix or Decision Matrix can be used to narrow down the available alternatives.
    3. The outcomes, which are usually probabilistic, are the resulting situations that arise by pursuing the selected alternatives.
    4. The payoffs or benefits are the values placed on the outcomes associated with each alternative. The payoff values are a combination of tangible and intangible benefits.These four elements of the decision problem (objectives, alternatives, outcomes, and payoffs) should help the business select the best or optimal alternative to implement the decision using the appropriate decision science techniques.

These four elements of the decision problem (objectives, alternatives, outcomes, and payoffs) should help the business select the best or optimal alternative to implement the decision using the appropriate decision science techniques.

Elements of a Decision Problem

Figure 2: Elements of a Decision Problem

3. Once the decision to implement the alternatives is made, the next step is to identify the resources needed to execute the decision. The resources could be time, skills, budget, equipment, and even data. The key step in executing the decision is to manage change with the right ownership or accountability. Successful change initiatives are often associated with strong accountability or ownership. This means having an accountable leader who is close to the objective and KPIs being tracked for business performance. For example, if the KPI is on “Days Payable Outstanding (DPO)” to improve the cash conversion cycle (CCC), it is advisable to have the Account Payable (AP) Manager track and improve the DPO KPI. Basically, identifying the right and competent leader will help in mobilizing the necessary support, including the desired resources and culture, to convert the insights and decisions into action.

The effectiveness of a decision can be validated with appropriate feedback mechanisms given that drifts in models and data are bound to happen as the business  evolves and adapts to change. The best way to manage model and data drift is by continuously measuring the performance of data and models using the right KPIs [Southekal, 2021] and feedback mechanisms.  The insight value chain or the process flow and its components discussed above are shown in the figure below.

Insights Process Flow

Figure 3: Insight Process Flow

Today, actionable insights are at the heart of every business decision to help companies increase revenue, reduce expenses, and manage risk. The purpose of insights is not just to know; it is also to know and act thereby make informed decisions and improve business performance. The three steps described above can help organizations formulate the right business and data strategy for turning data into performance insights and then into actionable insights for improved business results and performance.



This article was written by Prashanth Southekal, the Managing Principal of DBP Institute (www.dbp­, a data and analytics consulting, research, and education firm. He is a Consultant, Author, and Professor. He has worked and consulted for over 80 organizations including P&G, GE, Shell, Apple, and SAP. He is the inventor of DEAR Model, a systematic and structured approach for data-driven decision-making. Dr. Southekal is the author of two books – “Data for Business Performance” and “Analytics Best Practices” – and writes regularly on data, analytics, and machine learning in Forbes, FP&A Trends, and CFO.University. Apart from his consulting pursuits, he has trained over 3,000 professionals worldwide in Data and Analytics. Dr. Southekal is also an Adjunct Professor of Data and Analytics at IE Business School (Madrid, Spain) and CDO Magazine included him in the top 75 global academic data leaders of 2022.  He holds a Ph.D. from ESC Lille (FR) and an MBA from Kellogg School of Management (U.S.). He lives in Calgary, Canada with his wife, two children, and a high-energy Goldendoodle dog. Outside work, he loves juggling and cricket.


The Practical Guide to Using a Semantic Layer for Data & Analytics
Semantic Layer - diagram

Request a Demo