Customer service problems that look simple rarely are. Consider an unexpected charge on a bill. The customer wants the issue resolved immediately. But the service rep needs to review the account history in detail to investigate. As the customer waits for a resolution, trust slowly erodes.
AI is already helping sales and service teams navigate some of this complexity to deliver better experiences. Now agentic AI can take on more of the reasoning and coordination involved: interpreting records, analyzing the customer’s intent, and orchestrating steps to resolve the issue. That raises a fundamental question: What should an AI agent be trusted to do on its own, and how much authority should it have?
Deciding where, and how, to introduce greater autonomy into customer relationship management (CRM) processes can feel overwhelming. The customer journey spans many parts of the business, from sales to fulfillment to service.
We talked with ServiceNow partners about how they’re helping customers modernize CRM. Our conversations revealed three questions that can make the process more manageable.
You can’t transform everything at once. Start by defining the problem you want to solve and the outcome you want to achieve. Then look beyond individual tasks to understand the workflow surrounding it and evaluate where autonomy could make a meaningful difference.
Konversational, a customer experience consultancy and ServiceNow partner, recommends starting with the business outcome you want to achieve. Many of its clients are trying to solve similar challenges:
- How do we get invoices out the door faster?
- How do we track people's time to ensure we’re properly billing customers?
- How do we onboard new clients quickly to get to cash faster?
“An onboarding process might have 180 steps that are highly manual,” says Andrew Paolino, general manager of sales for the Americas at Konversational. “We look at the entire workflow to determine which steps can be fully automated. There isn’t one right answer; every customer’s journey is a little different.”
The objective isn’t autonomy; it’s a better outcome. Be clear about what you want to achieve and how much autonomy can help you get there.
Once you’ve defined the problem, you can work backward to determine which data, systems, and workflows need to connect. The key to success is understanding which things to connect, not connecting everything.
Autonomy requires context. AI needs to understand the customer, the work, the rules, and the actions that happened before it can take an appropriate next action.
CDW, an IT solutions provider and ServiceNow partner, stresses the importance of designing workflows with AI in mind. Equally important is having clarity on how data moves from one step of the process to the next.
“When a customer calls, your CRM solution needs to route the call to a team member who has the right context to help,” explains Jimmy Schmitzer, a business strategy architect at CDW.
“More importantly,” he continues, “the AI solution behind it needs relevant context to do proper inference and reasoning. You need to look at the technology that sits at each layer and figure out the synergies to make sure there’s a clear, concise, seamless motion across the customer journey.”
When the right signals are connected, autonomous CRM capabilities can help you anticipate issues and determine what should happen next. Aria Systems, a cloud billing provider and ServiceNow partner, explained how that can change the dynamics of a common customer complaint.
“Instead of waiting for an incoming call about a billing error, autonomous agents help prevent that call altogether. They proactively flag the issue to the customer, resolve it, and can even suggest another service or product that makes sense,” says Brendan O’Brien, chief innovation officer and co-founder of Aria Systems.
“But to make good recommendations, agents need access to contextually appropriate data—account history, escalation history, prior service interactions, and billing information,” he adds.
The better the context AI has, the more confidence organizations can have in its actions. But context alone isn’t enough. You also need visibility into the decisions AI agents make and the actions they take, along with appropriate oversight. That visibility and governance need to be designed into agents before they’re deployed, not added later.
The value that CRM modernization delivers looks different to every organization. The best measure depends on the problem you’re trying to solve.
Some process improvements deliver tangible results, such as accelerating response time or order fulfillment. ServiceNow partner Everforth GlideFast changed the way a bioengineering company moved assets between customer sites.
“We built a new app on ServiceNow to help mass-transfer assets,” explains Josh McMullen, associate vice president of CRM and industry professional services at Everforth GlideFast. “A site agent and site worker can pull up the app on their phone and migrate assets with the click of a button, taking 20 hours of manual labor down to about five minutes of automated work.”
As organizations deploy automation across more complex work, the measures of value can expand beyond efficiency to the quality and outcomes of the decisions agents are making.
Xactly, a sales performance software company and ServiceNow partner, explained how AI agents can be used to help sales teams meet their goals and move deals forward. An AI agent might:
- Flag that a sales team will fall short of projections
- Identify deals in the final stages of closing
- Recommend ways to address the shortfall before the end of the quarter
AI agents could also help sellers make more strategic decisions during the last mile of a sale.
“If a rep configuring an order is offering a larger discount than necessary, an agent could show how adjusting the discount to 10% would improve the deal and help the rep reach a threshold in their compensation plan,” says Jim Kirby, vice president and head of global partnerships at Xactly.
“Instead of simply trying to get the deal done by Friday, the seller can make a decision with a better understanding of trade-offs, operating more like an owner of their business,” he adds.
When priorities and outcomes are clear from the outset, it’s easier to measure and build value over time.
CRM touches nearly every part of the customer journey. The good news is you don’t need to automate everything at once to make a meaningful difference. Identify a specific point of friction, examine the workflow holistically, and determine the context required to help eliminate it.
Keep the ambition big even when the starting point is small. The real value of autonomous CRM comes from rethinking how work gets done, not doing the same work faster.
Find out how ServiceNow can help you break free with autonomous CRM.
