A distinguishing of this article is its care to place the human element in the center. Examples of the kind of care I'm talking about include human judgment, management teams' collective knowledge, interpersonal communication, the cultural gap between managers and data analysts, and people's propensity to believe the hype and inflated expectations about new technologies at the expense of reality. Similar to this, the use of big data, analytics and artificial intelligence (AI) is carefully discussed in plain terms.
To understand this article, you don't have to be a technologist.
To understand this article, you don't have to be a technologist.
Before we start, it is crucial to understand the profile of the business decision-maker. We assume managers at all organisation levels are key business decision-makers and those who work with data and analytics usually should support them.
But when managers hear about big data and analytics, what do they think? Do they comprehend how these technologies can be applied to decision-making?
Director of Data and Analytics: "Even though some managers completely grasp it, but not everyone is as engaged. And the majority of business people believe that analytics and data are challenging and that they require assistance. And unless we provide them with data analytics, it is useless to consider how they fit this into the business.
CEO: I believe that many of the difficulties are related to education. Data can imply many things to many people but at its most extreme, from ”I use that to recharge my phone” to “Oh, I completely comprehend the value of data, and I am aware of the differences between managing data, employing tools to analyse it, and gaining insights.”
Business Analyst: "Sometimes I sense that the CEO's instructions to review the data again are coming from someone who is afraid, suspicious, or misinformed about what the data actually means. In that scenario, I might need to reemphasize or sell the argument more effectively, as well as perhaps do a better job of describing what the data means. On the other side, the CEO will grasp the business and the data if he or she has an analytical background. He understands all sides, so if he wants it, it's not just because he's the CEO.
Chief Data Officer: "Knowing the appropriate questions to ask is the real issue, and those questions should be based on your knowledge of the value that data can add to your company. I therefore frequently see businesspeople who state, "I have a need or an issue, and I believe that data can assist," to which I reply, "Okay, let's chat, describe what you need.”
But from another side of the equation, we have managers, who are real decision makers at all organisations levels from strategic to operational business decisions.
Technologists and managers are educated and trained to think and communicate about technology and business differently. The gap caused by the disparities between technologists and managers needs to be minimised in an era where data-driven decision-making is emerging as the new standard and where technology and business are converging.
The Different Mindsets of Managers and Data Experts
Due to their varied backgrounds, experiences, and training, managers and data professionals have different mindsets—different approaches to comprehending and solving problems. To take advantage of the capabilities brought by these varied mindsets, it is crucial to comprehend the distinctions between them. Thinking at these distinctions in terms of computational knowledge and contextual knowledge is one way to approach the topic. High degrees of computational knowledge are typically possessed by data professionals who specialise in the collection, analysis, and interpretation of data. On the other hand, managers will typically possess higher levels of contextual knowledge, which, depending on their level of education (an MBA, for instance), may include knowledge of finance, human resources, economics, and marketing, in addition to their specific areas of domain experience and expertise.
Most managers are unlikely to have extensive computational expertise unless they have prior technology experience or education. In the end, the manager's contextual expertise determines how well the data is turned into knowledge and how well the analytics results are applied to the decisions. It's crucial to remember that data-based decisions, particularly those influenced by big data and analytics, are technologically bounded.
These mindset differences create data analytics bottlenecks within the organisation. A lot of managers are hungry for data to back up their business decisions, but they are not technology-bound enough to get these analytics themselves because data experts capacity is limited mainly to support TOP managers with strategic business decisions.
Thus typical managers usually make business decisions based on their intuition or trust without required deep data analytics. A 2018 survey of New Zealand managers (from directors and board members to supervisory-level managers) found that 95.7 percent of these managers were familiar with the term ‘analytics’ and 89.6 percent were at least moderately familiar with the term ‘big data’. The study also found that almost 60 percent of managers often or always relied on outputs from data analytics for decision-making. All of the managers mentioned that they at least sometimes incorporate their own intuition and experience into their decision-making. At the same time, however, the study found that a quarter of the managers said they value or trust their intuition and experience more than analytics, while about the same number of participants (28.4 percent) said just the opposite. The study also found that those who favored analytics over intuition were more often mid-level managers who were not in a position to use big data insights for strategic company decisions. Top executives were generally not as competent as they could be in using analytic tools and techniques and seemed to rely on other managers within the organization to generate big data insights, and those insights are then used to confirm their own intuition or are ignored if they conflict with their gut feeling.
Computational and contextual knowledge should make a powerful combination when it comes to solving problems and making decisions, but they must be in sync and complement each other.
The evolution of data analytics provides managers with endless numbers of dashboards with a significant increase in available data. For contextual employees, it might be even confusing as a result of a lower understanding variety of data on dashboards. However, row data alone can’t assist in improving managerial decision-making. To gain insights from data, it has to be processed in a timely manner and put into a form that will be useful for managers mindsets. It is self-service analytics tools, that can turn data into analytics from managers questions based on their vision and needs.
Managers then can use this information in combination with their knowledge, experience, and insights to improve their decision-making.
Similar to google search but with a chat interface, which can ask additional questions in case of ambiguous meanings based on complex variations of enterprise databases. Self-service Analytics tools can detect ambiguous notions and double-check each of them with additional questions to the managers in order to provide the most accurate calculations that fit the personal knowledge model of the end user.
Self-service analytics tools features usually include:
Diagnostic analytics tries to answer the question “why something had happened?”. Generally, statisticians and a variety of data visualisation techniques are used to answer this question. This type of feature explores the root causes of incidents using historical data. With diagnostic analytics, managers can drill down into data with their questions to find out the root causes and gain new insights about any problem before making decisions. This drilling down can be through more questions or the use of interactive visualisations on most types of reports.
Predictive analytics enables managers to make more prudent and forward-looking decisions since the built-in statistical models are designed to predict future conditions and notify managers in Microsoft Teams/Slack App if something went as not expected. Predictive analytics utilises quantitative and qualitative techniques and can forecast various scenarios based on supervised, unsupervised or semi-supervised machine learning models. Data and mathematical techniques are used to discover explanatory and predictive patterns, which represent the inherent relationships between data inputs and outputs.
Prescriptive analytics provides the decision maker with sufficient information about prices and estimated volume, ROI, costs of production and EBITDA at any level to clearly determine the best course of action. Empirical evidence from a study on optimised organisational sales history can suggest that revenue for specific products can be increased marginally.
Descriptive analytics serves the purpose of well-defined past and present opportunities. The real-time information gained from Self-Service Analytics tools enables managers to alter or adapt their future behavior and improve negotiation skills in front of customers or suppliers.
Using self-service AI analytics assistants can help not only with data analytics but with best practices sharing between experienced colleagues and newbies by leveraging the questions history analytics and relevant prediction of the most useful question for undiscovered insights for new employees. It significantly improves the learning curve for new colleagues by sharing relevant KPIs and deep data insights straight away.
While many scenarios require straightforward decisions from managers (such as poor sales, employee low performance, etc.), other situations requiring decisions and action might not be as obvious. The vast amounts of data and information that organisations produce and that managers must sort through may hide it as part of the "status quo" or they may be buried in its analytics. Managers and business analysts must be able to actively seek out decision-making opportunities in order to be effective given the complex and rapidly changing business environments and shifting stakeholder expectations. They must also continuously strive to meet both present and future expectations and conditions. This requires managers to have self-service analytics tools for helping themselves in both finding and solving business problems.
Erik Brynjolfsson's empirical research backs up the beneficial effects of data use on organisational performance. According to their research, data-driven decision-making boosts a company's productivity by 5–6% and has an impact on market value, asset utilisation, and return on equity. According to a different study, businesses in the US manufacturing sector that used data-driven decision-making generated around 3% more value. These companies showed improved performance and increased output. Another empirical study supports these findings, demonstrating that self-service analytics tools have a direct positive impact on information processing power, which in turn has a favorable impact on data-driven decision-making and increases overall decision-making effectiveness.
Real-world decisions are influenced by both the situation's possible external constraints and the decision-own maker's emotional and cognitive experiences, such as gut instincts, intuition, insight, and judgment—all of which are referred to non-rational modes.
We now reach the topic of how technology affects judgment. The majority of technology explicitly supports rationalising decision-making. In the belief that more data and knowledge will lead to more informed decisions, self-service analytics tools like NLSQL generally help in providing data and information to the decision-maker.
When used properly, rational decisions produce outcomes that are testable, repeatable, and supported by data and analytics. Non-rational decisions may be justified and frequently derive from years of experience, but they are more subjective, which may restrict their ability to persuade other stakeholders. When decision-making circumstances are straightforward, as with many operational decisions, rational conclusions are easy to make using self-service analytics tools directly at corporate messaging applications.
For analytics to be used effectively, the proper question must be asked and the prerequisites must be specified. These inquiries deal with the motivation or purpose behind the analytics effort, the metrics that must be established for the data and analytics to be used effectively, and the desired outcomes. The proper question must be asked in order to begin the analytics process, and asking the right question is seen as a crucial decision-making skill.
”Start with the question, not the data” is an innovative approach that becomes possible for non-technical employees to leverage the power of big data with self-service data analytics tools.
Both human judgment and analytics can identify a decision situation, which may be either a problem or an opportunity. These roles should work in tandem to strengthen or challenge initial decisions regarding the situation.