Building Deep B2B Customer Acumen with Human-in-the-Loop AI

0
128 views

Share on LinkedIn

Customer service chatbots, intelligent avatars and the rest of the Artificial Intelligence (AI) crew are no doubt heading our way, poised to alter the customer experience (CX) forever. But AI already is impacting business-to-business companies in less hyped, but equally dramatic ways. It’s helping them add depth and dimension to their customer interactions – and save significant amounts of revenue – by building a full 360-degree view of their customer relationships.

The help is sorely needed. Like any other business, B2B organizations know they need to cultivate intense sensitivity to customers’ needs and motivations to thrive in today’s markets. But they’re struggling. The B2B world is document-heavy. It’s dominated by contracts and their webs of amendments and supporting documents, all of which are encased in legalese and specialized formats that resist automation. It’s also systems-heavy, with important customer data scattered across a jungle of technologies – everything from contract repositories to billing systems – that don’t talk to each other well, or at all. As a result, deep customer acumen is often elusive for even the most customer-centric B2B enterprises.

The Quest for Customer Insight

To access the information they need, reps go on time-wasting data hunts through a tangle of documents and systems. Consider what might seem a straightforward task – figuring out the date when a contract is up for renewal. The contract may state the date, but more likely you have to calculate it from a renewal term – say a year from the date the contract was signed plus an advance notification period. Or it may depend on information that’s external to the contract, such as a delivery date, which you need to pull up from a different system. It’s a similar story for many kinds of customer data – pricing agreements, purchase histories, credits and discounts, warranties… it’s a long list.

The effect on the customer conversation can be jarring. Even simple customer inquiries are met with an “I’ll get back to you.” Time is lost, inertia sets in, and before you know it you’ve lost a buyer. The same complexities and inefficiencies hamper planning and strategy. It’s hard to design a solid go-to-market strategy when you have trouble aggregating even basic information, like which customers have bought which products.

Companies are losing more than the opportunity to demonstrate commitment to service by providing smooth, fluent customer interactions. They’re losing revenue. Getting a contract renewal date wrong could mean missing a chance to cross-sell. Overlooking the end of a negotiated discount period could be costly. B2B companies not only have an opportunity to provide a better CX, but they can also demonstrate the value of that effort in no uncertain terms – in an actual dollar-amount impact on the bottom line.

The Human/AI Mind Meld

This is where artificial intelligence comes in, especially the powerful approach called human-in-the-loop AI.

Why the human component? B2B companies know that their customer data contains valuable insights, but finding those insights with AI algorithms alone is like looking for a needle in a haystack with a magnifying glass. Knowing the output you’re looking for from the start is critical. Human expertise can provide the context that’s needed to conduct a controlled, targeted search.

More importantly, human expertise can ensure that the input data is accurate and up-to-date. The old garbage-in, garbage-out maxim applies just as much to AI as to any other IT system. You can’t just throw massive amounts of low-quality data at AI and expect to get great insights. With the human touch, you can achieve upwards of 99 percent accuracy to improve the AI’s training sets in short order.

Human-in-the-loop AI may seem like heavy-duty machinery to bring to bear on customer experience, but in reality the human element makes this approach light on its feet and highly maneuverable – much more so than a big data/analytics approach. Instead of spending years building up massive data lakes, you can start with small data sets at high accuracy and quickly stack up a series of wins that strengthen the customer relationship and close revenue leaks at the same time.

To go back to the renewal date example, the human-in-the-loop AI platform can identify upcoming renewal events for, say, the current quarter, prioritize them, and push them out to the people handling the relationships. Your renewal teams are notified by the system well in advance so they have plenty of time to prepare, they clearly understand the customers’ needs, and they’re aware of the areas where it would be appropriate to make new offers.

Maybe this approach can’t match the glamour of the chatbots and the virtual agents, but it demonstrates the power of humans and AI working together, each doing what they do best in pursuit of a clear objective: a deeper understanding of customer relationships and the value they bring to businesses.

LEAVE A REPLY

Please enter your comment!
Please enter your name here