AI is officially a business priority. IDC estimates that $35.8 billion was spent on AI in 2019 a 44% increase from the previous year, and 1,500 enterprise leaders told Accenture that they are worried they will go out of business in five years if they fail to scale AI. Even businesses that have implemented AI are not seeing that promised ROI. Research by Gartner, IDC and MITSloan Management Review & Boston Consulting Group all found that businesses are struggling to generate a positive return on the advanced algorithms we call AI.
So, is AI just a lose-lose for businesses? Simply put, “no.” Companies are struggling with AI because they are taking half measures, only implementing them in small use cases or in environments where the data is insufficient to power its success. That’s causing their businesses to flounder.
Enterprise AI isn’t something that businesses can dip their toes into. Companywide buy-in and methodical planning ensure that one’s eyes aren’t too big for his stomach. That can be difficult when economic expectations are high. Afterall, Statista estimates that the AI software market could be valued at more than $118 billion by 2025, and according to PwC, AI has the potential to contribute $15.7 trillion to the global economy by 2030 and boost GDP for local economies by as much as 26%.
Yet, return on investment is directly correlated with the sophistication of the AI deployment. Successfully implementing AI requires that companies think big—and with long-term goals in mind.
Operating a 24/7 business
Take, for example, a business that needs to operate 24/7. There are many challenges and expenses associated with round-the-clock, particularly multi-market service. The biggest one is the struggle to acquire enough talent to be in the office at all times. That’s why most businesses limit many services within a single time zone. But in this constantly connected world, where clients and customers expect assistance at any moment, the need for an always-on solution has never been more pressing.
Cognitive AI—that is AI capable of “learning” over time—is able to turn any business into an always-on and always ready entity. By drawing on organizational data, cognitive AI is able to answer customer questions around the clock and handle tasks like form generation, with little to no human involvement. This allows customers from around the world to engage with the business whenever it is most convenient for them and substantially reduces the strain of providing 24/7 service on the human employees. It’s also scalable. Telefónica, the multinational telecom, first used digital voice assistants as whisper agents in their call centers. Today, an intelligent voice assistant fields 4.5 million calls per month, or 100% of all calls around the world, day or night. By identifying customer call abandonment as a key issue and tackling the problem with a cognitive AI solution, Telefónica was able to use AI at scale, saving money and improving customer experience in the process.
Ultimately, AI backed by machine learning and neural network technology stands to revolutionize the business world by allowing the automation of cognitive processes, encompassing the core of decision-making, learning and conversational engagements.
The most advanced cognitive AI can even find ways to complete various tasks and learn “on the job,” which translates to less human intervention and upkeep. In contrast, a more simplistic solution like an RPA bot would need to be reprogrammed with each new learning, where a cognitive AI solution will simply adjust to the environment.
The above example is merely one case where AI can be used to not only drive positive ROI, but as a lever to change the way business is done. This however is by no means the only way to use AI. According to McKinsey, 45% of workplace activities which currently rely on human intermediation could be replicated by machines and therefore, be executed with machine efficiency. That isn’t to say that humans would be replaced, however. Humans will still be central to identifying the tasks necessary to achieve a result, and they would be responsible for building the workflow based on those tasks, but AI can build the automations that transform how business is done.
Previously, the only way to create automations that addressed a company’s unique needs was to have a dedicated team of engineers research business processes and then identify and describe all necessary steps to reach a resolution. In contrast, automation teams could consist of engineers who are educated in specific business processes along with training algorithms that allow them to do the work in significantly less time. This process is still used in many enterprises, but the problem is, the automations they spend time implementing are rarely instances of “once fixed, forever fixed,” and often require an iterative follow-up process.
With intelligent automation, however, systems can combine analytics, cognitive AI and guided machine learning to hasten the creation of new automations. Humans are only necessary at either end of the process: at the start, they would set high-level directions for the system (e.g., giving the directive to observe a specific task like a hardware procurement workflow and then recommend automations based on those observations). At the end, they would approve suggested automations before they go into production. Without having to execute the entire process themselves, the human staff can spend more time on the work that adds value, such as conceiving new products and services or improving the customer journey.
Making the right investment
It is not an exaggeration to say that AI will revolutionize the enterprise and bring a wealth of financial and productivity gains. Its promise makes it an unavoidable line item for any business hoping to compete in the global economy. But, like any investment, if it is poorly implemented, it will be high-risk, costing companies so much in the long-term that they may have to shut their doors. Making the most of this evolving technology will require looking beyond point automations and looking instead at automating processes that are key to far-reaching business goals. That’s the kind of thinking that will allow businesses to gain efficiency, save on overhead and become more agile. But everyone must be willing to dive in and embrace dramatic change.