Messaging, Chatbots, AI: Finding the Enterprise Opportunity
The natural fit is in providing a consumer-like customer experience, but first companies must address the 'contact center paradox.'
We hear a lot these days about how the digital generation prefers messaging to voice, and we're starting to hear a bit of the same regarding chatbots and artificial intelligence -- especially for applications in the consumer world. While messaging, chatbots, and AI each have a role to play for enterprise users as well, developers do face a challenge in finding the right fit in there.
A colleague and I recently presented a session on messaging, chatbots, and AI at Jeff Pulver's Spring 2017 MoNage, a conference focused on the future of the conversational Web, chatbots, and messaging. Many developers attended, and during the conference we saw great innovation, both from those that have launched successful apps and those with promising applications in development. Not surprisingly, the focus today is mostly on consumer apps, but I did find some tie-ins to contact centers and digital customer service. This is to be expected given that so much of the work today around chatbots and AI is about driving online commerce.
Identifying the Opportunity
During our session, we explained the realities of selling these emerging technologies into the enterprise environment. I've written about this elsewhere, with the main message being to focus on addressing real business problems. Chatbots sound cool, and AI sounds intriguing, but where and how are they going to drive business value?
Since the business value of these technologies continues to evolve, we talked about the big challenges developers face in selling chatbots and AI into the enterprise. The challenges may be daunting, and very different from selling these technologies into the consumer world, but the opportunities can also be large... so the potential payoff is definitely there.
UC and collaboration certainly represent one such opportunity, especially around automating scheduling tasks and managing workflows. However, this type of automation primarily relates to productivity, and while improved productivity is important, measuring impact can be difficult. This means UCC vendors must demonstrate how their solutions truly impact higher-order priorities, namely reducing costs, growing revenues, and managing risk, all of which are easier to quantify.
The contact center, which has already emerged as a focus for many developers, serves as a good example for how to position chatbots and AI for the enterprise market. The first step is to identify the problem, and in this case, the opportunity is clear -- to improve customer service in order to reduce the high cost associated with poor customer service. As I cited in the session, industry data shows how unhappy customers will buy less and/or go elsewhere, while happy customers will keep buying and are more likely to recommend a company to others.
The Contact Center Paradox
The high cost associated with poor customer service alone will get the attention of enterprise decision makers, but the use case becomes stronger when drilling down a bit to focus on the real challenges facing contact centers, and why that's giving rise to unmet customer expectations. In today's globalized always-on market, customers expect 24/7 service, but they also want personalized service to feel valued and to get the desired outcome.
I call this the "contact center paradox" because it's an impossible balance to strike given the higher priorities cited above. The cost of providing this level of service with live agents is prohibitive, so something has to give. Customers can't have it both ways, so contact centers are experimenting with chatbots to bring their costs down, but also to maintain enough personal touch via automation and self-service.
Contact centers know they need to make more use of chatbots and AI if they are to catch up to consumers in terms of technology adoption. The gap between the experience delivered and the experience expected plays a big role in why customer satisfaction ratings continue trending downward, particularly when legacy systems are in use. However, many customers still prefer to deal with live agents, and are not receptive to automated alternatives -- and that puts contact centers in a tight spot.
One issue is that chatbot and AI have a long way to go before they're ready for widespread adoption. These tools will take time before they can provide real value and earn the trust of customers, so companies that adopt them do have to consider the risk factor. For example, speech recognition accuracy is approaching 90%, but is that good enough for the contact center? Skeptics will point to Microsoft's early experience with its Tay chatbot, and flat out say, "No."
Another issue is the concern about chatbots putting agents out of work. While there might be some truth to that, contact centers should view chatbots as complementing the work of live agents, taking the simple tasks off their plates, and enabling them to optimize the personal touch, which matters most.
Technology is definitely changing the contact center space, and developers need to understand what the real value drivers are going to be to have success selling into it. As I mentioned during my presentation, data shows a positive response to actual experiences with chatbots for basic customer service, and that such interactions largely met expectations based on perceptions of what customers thought the experience would be like. As such, while the shortcomings of automated service are easy to find, progress is being made, and it would be short-sighted to dismiss its potential for the paradox I've been describing here.
What to Do
Contact centers can't afford to wait for speech recognition accuracy to reach 100% -- if it ever does -- to deploy it more boldly. Customer expectations are evolving too quickly to wait for the perfect time, and companies need to be willing to try new things. And certainly the need for contact centers to find a better way is urgent.
The key will be for contact centers and developers to accept some risk to refine these applications, and let these technologies find their legs. Clearly, some customers are ready for this now, and that's where the focus should be.
The contact center paradox is real, and the challenges will only get worse unless companies take steps to close the gap. This is a problem enterprise decision makers understand, and it represents a great entry point for developers looking to grow beyond the e-commerce space. Given the impact contact centers ultimately have on driving sales, it's fair to say that the developer community knows a thing or two about that as well.