The nature of business decision making is changing. Improving the accessibility of data may seem like a positive development for reaching well-informed conclusions, but many leaders struggle.
For example, a Signal AI survey of more than 1,000 executives found that the most common barrier to decision-making was “overwhelming amounts of data,” with 44% of respondents finding data difficult to parse. is reported to be
But the solution to modern leadership demands may lie in augmented decision-making. Here, artificial intelligence (AI) or machine learning (ML) suggests or assists in reaching a result.
For example, Publicis Sapient, a digital transformation consulting firm, has made AI and ML technologies an integral part of its client-facing services and internal operations. The company has developed models to predict employee turnover, manage revenue, and combine customer, website, and third-party data to identify marketing and sales opportunities. Nigel Vaz, his CEO at the company, says the advanced use of data to inform business strategy will improve the performance of many companies.
“The most successful organizations today, and certainly tomorrow, are those that either start as data-driven or transform to become data-driven, with data-driven leadership at their core,” says Vaz. “As AI and machine learning technologies evolve, leading companies will feel the need to transform how they work to deliver value.”
How to prepare for the rise of augmented decision-making
Vaz adds that his leadership team uses augmented decision-making to “drive more informed, more data-driven conversations.” This process examines macroeconomic trends, business indicators and client needs to identify opportunities. Vaz attributes this approach to the company’s record growth of 19% in the past fiscal year.
Leaders must ensure data is sanitized from known discriminatory practices that can skew algorithms
This technology has the potential to completely change the way executives make decisions. Professor Chris Tucci teaches the ML and AI Executive Education Program at Imperial College London Business School. He believes the underlying model is likely to become more “creative” in the years to come.
“AI systems will soon be available for brainstorming ideas for developing new markets and customer segments, new products and processes, and new business models. It could be a kind of ancillary to human decision makers,” says Tucci.
Using data processing tools in this way requires a significant change in the attitude of certain leaders. Management surveys repeatedly show distrust of analytics.
In one such report from Deloitte in 2019, 67% of executives reported being “uneasy” about accessing or acting on data from advanced analytics systems. increase. In 2018, he said the percentage of his CEOs who responded to KPMG followed their intuition for data-driven insights.
However, there are signs that the tide is turning. “Executives have resisted data analysis for higher-level decision-making. said John Hill, founder and CEO of Silico, an AI-powered platform that enables businesses to simulate business decisions and processes. “But now more and more executives are betting their business entirely on AI and ML to power critical business decisions and plan autonomously.”
Overcoming the main barriers to AI adoption
The promised results are appetizing. Signal AI, for example, estimates that US companies missed $4.26 trillion (£3.43 trillion) in revenue in 2020 due to slow adoption of augmented decision-making.
But hiring may not be so simple. Hill warns that companies looking to capitalize on this opportunity may not necessarily have the infrastructure or capabilities in-house and risk reputational damage if the data is misused. I’m here. “The most common barriers to adopting augmented decision-making include the difficulty of interpreting results from AI, the difficulty of integrating AI with existing systems, and concerns about data privacy and bias. concerns,” he says.
“Assure executives that data has been sanitized from known discriminatory practices that may knowingly or unwittingly contain implicit biases and may skew algorithms. For that reason, we need to scrutinize the data used for higher-level decision-making,” he explains.
If you have fiduciary or ethical responsibility, you should always understand how decisions are made.
Further problems can arise when leaders use “black box” technology (AI-powered algorithms that do not reveal methodology) without proper care and understanding.
Tucci doesn’t necessarily think executives should avoid these programs. All use cases so far have required humans to control input and scrutinize final decisions. However, Vaz argues that the leader must have some knowledge of how the model works, even if humans are still acting as the controlling variable for the process.
“It is important that business leaders understand the fundamentals of AI and machine learning and how these technologies affect decision-making processes. always need to understand how decisions are made,” says Vaz.
“The reality of today’s tools is that they need a prompt to initiate recommendations,” he continues. “So part of the process of creating solutions is how to provide effective prompts for individuals using these tools to shape their output.”
The era of hybrid management?
One of the inevitable consequences of the rise of this technology is the question of how augmented decision-making will change the composition of management itself. The last few years have already seen a surge in the number of Chief Data Analytics Officers (CDAOs). According to NewVantage data, 77% of Fortune 500 companies now have a CDAO, compared to 12% in 2012.
For other positions on the leadership team, Vaz believes technology should take some of the burden off the CEO and others. Instead, decision-making responsibilities are shared across the organization, leveraging the agility provided by AI and ML processes.
“Augmented decision-making represents a new way for data to move through an organization to make better decisions at every level,” says Vaz. “As a business leader, this is not about how these tools will change the way we make decisions as individuals. is.”
Promotion within an organization, and ultimately to senior management, can be based on the ability to understand and process data, provide the correct prompts to support programs, and deliver desired results for the business. I have. “You don’t have to be a coder or know how to write your own AI programs. says Tucci.
