Is the buzz around artificial intelligence justified?
Two-thirds of senior executives believe that AI is important for the future of their business, but the average return on all AI investments by company is still struggling to pass 1%. A new survey of more than 1,000 firms has warned that patience may be key to AI success, revealing that the majority of AI change programmes take more than two years to see a return on investment.
As the global economy faces headwinds from increasing import costs, trade wars and digital disruption – as well as the Covid-19 pandemic – many have been investing in artificial intelligence (AI) to help them to adapt to the difficult environment. Billed as a major opportunity in which employees can be redeployed from repetitive work to value adding activities, AI has also been said for years to be able to massively improve administrative accuracy, while reducing its costs.
According to analysis by Fortune Business Insights, the global AI market size is booming thanks to this hype, and was valued at $27.23 billion in 2019 and is projected to reach $266.92 billion by 2027, exhibiting a CAGR of 33.2% over that period.
New research from ESI ThoughtLab has cautioned executives against treating AI as a magic bullet to all their woes, however, and suggested that returns on investment usually take much longer to materialise than the average business leader might like to admit.
According to an examination of AI best practices, investment plans, and performance metrics of 1,200 firms, the majority of firms are posting positive returns on all AI areas. The area generating positive ROI for the largest percentage of companies is customer service, with 74% of respondents saying so, followed by IT operations (69%), and strategic planning (66%). With that being said, however, investing in AI is not a cure all.
ESI ThoughtLab said that 40% of projects are not yet showing positive ROI, based on an average ROI across all 19 areas. In fact, many firms advanced in implementing AI have yet to see positive gains. Underperforming areas include sales and business development, at 49%, and finance and auditing at 47%. The researchers suggested that this may be because businesses are underestimating how important the human side of digital change is.
Of the top performing firms in applying AI, 83% said they had been successful at developing, as opposed to just 9% of underperformers. In addition, overperformers were much better at training and enabling non-data-scientists to deploy AI, with 88% doing so, against 2% of underachievers. Illustrating how important this is in successful AI deployment, 61% of overperformers had decentralised their internal AI staff in some way, to help build AI teamwork across the firm, compared to just 22% of underperformers.
Even with the right approach, however, ESI ThoughtLab found that returns on AI investment do not always become pronounced quickly. While about two-thirds of senior executives believe that AI is important for the future of their business, the average return on all AI investments by company is still only 1.3%. Even the average return of overperformers of 4.3% pales against returns on other corporate investments, begging the question in some quarters as to whether the buzz around AI is still justified.
According to the researchers, the answer to this is still a resounding “yes it is,” but businesses will need to be patient when waiting to exit the payback period. An average of all firms suggests that the more familiar firms are with AI, the quicker it will pay off – beginners on average face payback phases of more than 1.6 years, while leaders will see this shorten to 1.2 years – however, this also depends on which industry an organisation resides in. For example, healthcare entities face the longest wait of 1.61 years, while the automotive sector averages a payback period of 1.26 years.
Commenting on the findings, ESI ThoughtLab CEO Lou Celi encouraged firms not to lose patience, as AI will become even more important in the coming months. Celi added, “As the pandemic propels businesses into a digital-first world, AI will become a key driver of corporate growth and competitiveness. But building proficiency in AI is not easy… It can fail to deliver results if the wrong business case is selected, the data is prepared incorrectly, or the model is not built for scale.”