3 reasons why executives fail in their artificial intelligence (AI) projects

How do most organizations start their journey to artificial intelligence (AI)?

Let’s take a look at how the executives of some large companies planned their foray into AI. Here are some recent examples from McKinsey:

  • The leader of a large company has spent two years and hundreds of millions of dollars on a company-wide data cleansing initiative. The intention was to have a data meta-model before launching any AI initiative.
  • The CEO of a large financial services firm hired 1,000 data scientists, each at an average cost of $ 250,000, to unleash the power of AI.

And here is an example that I witnessed firsthand.

  • The CEO of a major manufacturer has lined up a series of ambitious projects using unstructured data, as AI techniques are very effective with text, image and video data.

What do all of these initiatives have in common? They all failed.

In addition to the huge sunk costs incurred by these projects, they led to the organization’s disillusionment with advanced analytics.

McKinsey’s State of AI survey found that only 22% of companies using AI reported a significant impact on their bottom line.

It is not uncommon. McKinsey’s State of AI survey found that only 22% of companies using AI reported a significant impact on their bottom line. Why do so many projects fail and how can leaders avoid it?

[ Check out our primer on 10 key artificial intelligence terms for IT and business leaders: Cheat sheet: AI glossary. ]

Most executives who pursue AI miss three areas of ownership. These responsibilities start long before you plan your AI projects, and extend long after your projects go live.

Here are the three ways to defeat your AI initiative:

Mistake 1: Launching AI projects that don’t match the company’s vision

McKinsey found that only 30% of organizations aligned their AI strategy with business strategy. Isn’t it shocking that a majority of executives are burning their money in the name of AI? Organizations often pursue AI initiatives that seem worthwhile or those that are just urgent.

Organizations often pursue AI initiatives that seem worthwhile or those that are just urgent.

Certainly, your projects must solve a business problem. But, what is more important is that these results must align with your business strategy. Start with your business vision and identify how the data will enable it. Clarify who your target stakeholders are and define what success will look like for them.

Next, identify strategic initiatives that will empower stakeholders and bring them closer to their business goals. Now you are ready to think about the long list of AI projects worth reviewing.

In a report from the MIT Sloan Management Review, Steve Guise, IT director of Roche Pharmaceuticals, explains how AI is helping to transform the company’s business model. Roche strives to make personalized healthcare a reality. Guise emphasizes that the current model of drug delivery will not help them achieve this vision. They see the need to step up the pace of drug discovery from three to 30 drugs per year. Guise says AI can help them achieve this exponential improvement.

Roche is mainstreaming AI within the organization by strengthening capacities for screening, diagnosis and treatment. He’s increasing that by partnering with startups pursuing AI-driven drug discovery. Through these efforts, Roche has made significant advances in the treatment of diseases such as hepatitis B and Parkinson’s disease. By starting from their corporate vision and aligning all their AI initiatives with this overall goal, Roche’s efforts are paying off.

Mistake 2: Waiting to plan for ROI after the project goes live

When should you think about the return on investment (ROI) of your AI project? Most organizations make the mistake of tracking ROI when the project goes live. Executives are happy with fuzzy results such as “improved efficiency”, “brand value” or “happier customers” to make matters worse.

The leaders are satisfied with vague results.

Of course, it is not easy to quantify the monetary value of the results. But it is not impossible. You need to demand quantification of business benefits before you even give a project the green light. AI can generate value by increasing revenue or reducing expenses. Both are precious. Define which of these results your project will allow.

Identify a mix of advanced and lagging metrics that will help measure these results. Collect the data needed to calculate metrics by updating your processes or creating new ones. Finally, track your investments by going beyond the costs of hardware, software and technical staff. Include your spending on adoption and change management programs. This measure of ROI should be a critical factor in your project approval decision.

Deutsche Bank has rolled out its AI-based consumer credit product in Germany. The solution made a real-time decision on the loan even as the customer completed the loan application. Consumers were concerned that loan refusals would affect their credit rating. This product removed that risk by letting them know if their loan would be approved, even before they clicked “apply”.

Deutsche Bank found that loan issuance had increased 10 to 15 times in eight months after the AI-based service launched. The gains were achieved by attracting clients who would not have applied in the first place. This was a clear case of AI helping increase revenue.

Mistake 3: expecting AI-powered transformation without setting organizational culture

In its 2019 annual survey, Gartner asked Chief Data Officers about their biggest barrier to valuing analytics. The biggest challenge had nothing to do with data or technology. It was cultural.

As Peter Drucker said, “organizational culture eats strategy for breakfast”. Even the most well-developed AI strategy will be pointless if you don’t carefully shape the organizational culture. Culture change must start at the top. Leaders should use storytelling to inspire and demonstrate how AI can help the organization achieve its vision.

Leaders need to tackle the fear surrounding AI and improve data literacy for all employees.

Leaders need to tackle the fear surrounding AI and improve data literacy for all employees. They need to lead by example and support change by integrating data champions at all levels. Cultural change takes years and leaders have to influence it long after projects go live.

Wondering what is the main ingredient in Domino’s pizza? It’s data! Dominos Pizza is the emblematic child of technological transformation. The organization lives the culture of data-driven decision making and uses AI in sales, customer experience and delivery. This was not the case 10 years ago.

Patrick Doyle took over as the 50-year-old pizza maker in 2010 when he was criticized by customers and investors. Doyle made the bold decision to go public with the Harvest Reviews. He then performed a complete reboot in reverse and put the organization on the path to digital transformation. He made bold bets on technology by taking on risky projects, empowering people and creating several AI innovations in-house.

When Doyle retired in 2018, Dominos sales had grown for 28 consecutive quarters and produced better stock returns than Google. The outgoing CEO summed it up best: “We’re a tech company that sells pizza. By leading a cultural transformation within Dominos, Doyle ensured a transition to data-driven decisions that held true even after his transition to a new CEO.

How will you take AI beyond the innovation chasm in your organization?

Adopting technological innovation is never easy. Whether it’s bringing a new technology like AI to market or adopting it within an organization, the challenges are similar.

Innovators begin this journey within an organization. The innovation is then adopted by the first users, thanks to their initial enthusiasm and their openness to change. But then the pace slows down and sinks into an abyss. There is often a lack of visibility, uncertainty in results and a broader resistance to change.

This is where most initiatives fail.

For innovation like AI to cross this chasm and become mainstream, it needs leadership intervention. Leaders must ensure the success of AI by aligning the initiative with their business vision. They need to demonstrate economic value by institutionalizing conversations about AI ROI. Finally, they must shape the organizational culture to facilitate change and enable viral adoption of AI-based decision making.

[ Get the eBook: Top considerations for building a production-ready AI/ML environment. ]

Source link

Comments are closed.