Why patience is the key to unlocking AI’s full potential

It wasn’t long ago that accounting ledgers were written by hand and advertising campaigns were launched on highway billboards and in magazines. Back then, businesses didn’t have access to complex analytics, leading to major internal inefficiencies and knowledge gaps. Today, thanks to the power of digital and the growing sophistication of artificial intelligence (AI), all businesses are on an equal footing.

Across industries, companies are finding great traction for every rupee spent. With the added benefit of an infinitely rich data set for business intelligence analytics, digital has opened new floodgates of growth for businesses.

Consider the issue of customer data. When a typical business gets millions of new users every year, it becomes impossible to call each customer to determine the best prospects of converting them into bigger customers. Employing an army of sales personnel would be entirely impractical. This is where the power of artificial intelligence comes into the picture. AI engines have now evolved to be capable of converting large swathes of data through appropriate filters to gather business intelligence.

“What gets measured, gets managed,” said management guru Peter Drucker. While one application of AI is reactive, and helps perform quick fix remedies during the course of business, the most important aspect is its ability to predict. AI’s full potential can be unleashed when it helps in providing insights where human oversight may prove detrimental to the growth of the business. 

For example, if an international business is evaluating its web traffic around the globe, AI can spot traffic hikes from a particular region and predict the sales impact of this data. AI might suggest ways to capture and capitalise on this increased, region-specific interest. Business decision-makers can take proactive actions based on the data.

AI has become a mainstream aspect of business decision-making. Businesses are investing heavily in AI capabilities to solve problems in the short term. But AI can take years to bear fruit, gaining sophistication slowly and exponentially. It pays to be patient. With years of data fed into the engines, AI can, over the years, become extraordinarily efficient in migrating previously unmanageable amounts of data across various applications, doing more with less.

The power of voice

For businesses, customer retention forms the bedrock of their existence. Businesses should always be ready to provide information to the customer whenever it is needed, in whichever channel, and in whatever format is asked. Presently, most information is sought through Google searches, but it has become tiresome. Moving over to a live chatbot agent with an exhaustive knowledge base is an effective way to satiate modern customer demands and ensure that businesses are always there for the customers when they reach out.

In a way, computing is always about doing more with less and making it easier for users to interact with systems. However complex a system may be on the inside, user interfaces have been moving towards simplicity and personalisation. Leading the charge of change is voice technology. Humans respond well when using voice to interact with machines. With great strides in natural language processing and advanced engines, almost any language and accent has now become intelligible for machines.

Moving forward, in addition to chat or voice as popular channels, customers also expect a combination of the two. In many instances, a voice query can best be answered with a line response. A combination of voice and visual prompts will enable enriched and productive business conversations. In the years to come, such a multimodal approach to voice/chat/text customer interactions will help fix the issues surrounding contextual and mode-specific restrictions. For example, in the home environment, people would prefer voice as an acceptable interface, but in the business context, people just prefer to use voice as an authentication factor, and not as the primary input method. 

Productive AI implementation is all about balance. Too much automation can lead to customer backlash. Before going all in, businesses will do well to identify layers where human contact is still preferable to customers. Businesses should avoid excessive investment in areas where AI is unnecessary. If companies can pinpoint and pursue only the places where AI benefits users, they will have a tremendous leg-up in this competitive industry.

The writer is the ‘chief evangelist’ at Zoho. The views expressed are personal.

Raju Vegesna

---

Raju Vegesna is chief evangelist at Zoho. This views in this article are his own.