The field of customer service is a crucible of technology-driven change right now, and Guru is a company at the nexus of several of these trends. I recently recorded a podcast interview with its CEO Rick Nucci (embedded above), in which we explored the wide-ranging impact of connected digital technology on the customer service function.
I find Guru interesting because it's a company that connects up several strands that have always been kept separate in the past. Its primary function is to surface relevant how-to knowledge to customer service agents directly in their workflow. One of the strands therefore is how it keeps that knowledge fresh and maximizes its impact. This overlaps with a second strand, which is the increasing convergence of customer service with sales — in fact Guru in its marketing emphasizes its impact on revenue rather than merely customer satisfaction.
The final strand is the use of AI, not only to augment the effectiveness of call center agents, but also to measure that effectiveness over time and identify how it might be improved. This hits on a couple of themes that I'm big on. One of these is the virtuous cycle of engage - monitor - improve that I see as fundamental to the Everything-as-a-Service or XaaS model (pronounced 'X-ass'). The second is the emergence of a new conversational layer of interaction that allows AI to monitor human behaviors and measure their effectiveness. As I explained last week if you want to delve further, all of this fits into the wider framework of frictionless enterprise.
Our conversation was recorded on the fringes of the Slack Frontiers conference in San Francisco a couple of months ago, as Guru, along with several other vendors I'm watching, works closely as a partner with Slack.
Passive knowledge becomes dynamic
One 'ah-ha' moment that came during my conversation with Nucci was the realization that Guru is in the business of taking passive knowledge and making it dynamic. A core tenet of my philosophy of frictionless enterprise is the importance of sweeping away all vestiges of paper-based processes. Clearly, the notion of recording knowledge in a static database is a totally paper-centric concept that we no longer need to perpetuate in a digital world.
Knowledge should be inherently dynamic, because what we know constantly changes, and the subject of our knowledge also evolves — especially in a digital world where products gain new properties and functions in two- to three-month cycles — or even every day in a fast-moving DevOps environment. Guru not only keeps knowledge up-to-date, but also monitors and adjusts it based on how effective it is in resolving customer queries, as Nucci explains:
We are able to know and give back to our customers, the insights of exactly when you're using Guru and why. What is it in the chat that prompted the agent to use a Guru card? What was the email that prospect was sending to the sales rep that prompted you to use Guru knowledge?
What we do is, we correlate those events. We take those customer conversations, we correlate them with the Guru knowledge, and we learn and train our product, so that Guru can proactively bring knowledge to you ...
By having a product that our users will build daily habits around, what they're really doing is training the product, on behalf of their company, to be able to bring the best knowledge to folks when they need it, to have the best customer conversations.
Measuring and improving customer service
One important element of that is what Guru calls a trust score, which as Nucci explains, helps the vendor take action when its customer is showing signs of underperforming:
A trust score is the state of accuracy of [the] knowledge environment in Guru for a given customer at a given time. Our customer success team watches that very closely, and has signals set up that notifies you if that starts to drop down. We proactively reach out with the customer, we help with best practices, we diagnose what's going on.
The convergence of the customer service function into more of a sales role is particularly associated with the emergence of subscription-based contracts, where renewal becomes more of a focus for vendors. Making this work requires joined-up data, which is another area that Guru focuses on, by linking customer service performance to renewal success. Nucci tells me:
A customer service org ... traditionally will use metrics like, how quickly do we close the ticket? Now we're starting to look at metrics like, what impact are we having on retention for our customers? That's a totally different conversation.
Right now, your customer service leader is sitting at the table next to your sales leaders, where they're as important to the conversation — which is where they belong, in my opinion.
The winners in this next generation — of customer service not being a nice-to-have but as a must-have — will be the companies that rethink the way they measure and build their customer service orgs and recognize the need for empathy, but also how to tie what they do back to actual revenue in the company.
This notion of monitoring behavior with machine learning, recommending improvements, and then monitoring the results, is a new and important trend fueled by AI. As I wrote last week, Aera is another startup that's applying a similar approach to demand planning.
All of this requires some sophisticated gathering of previously siloed information and the intelligent application of AI resources. But I believe the companies that get it right will give themselves a big competitive advantage.