This article is sponsored by Cognigy and was written, edited, and published in alignment with our Emerj sponsored content guidelines. Learn more about our thought leadership and content creation services on our Emerj Media Services page.
Repetitive administrative tasks remain a significant source of employee burnout across various industries. In healthcare, as Microsoft’s Will Guyman pointed out on a recent episode of Emerj’s ‘AI in Business’ podcast, clinicians perform thousands of clicks per shift to manage documentation and non-clinical tasks, a workload strongly linked to burnout:
“Administrative burden is the first thing that comes to mind. It’s well known that clinicians spend a tremendous amount of time on administrative tasks and non-clinical care tasks. They do about 4,000 clicks per shift, and that really adds up in terms of burnout.”
– Will Guyman, Principal Group Product Manager in Healthcare AI Models at Microsoft
Similar dynamics play out in customer service, where agents routinely handle basic queries, such as password resets or account updates, diverting time and energy from more meaningful and complex interactions. Research by the Harvard Business Review notes that repetitive work is one of the main drivers of disengagement among employees. Automating these routine tasks through AI is becoming essential to improving both productivity and employee well-being.
A 2021 Salesforce study found that 89% of U.S. automation users reported greater job satisfaction, and 84% said they were more satisfied with their company as a result of automation—underscoring the value of reducing repetitive tasks through AI.
Looking ahead, the McKinsey Global Institute projects that activities accounting for up to 30% of hours currently worked across the U.S. economy could be automated by 2030—a trend accelerated by generative AI.
Emerj recently hosted a special series on the ‘AI in Business’ podcast, featuring Alan Ranger, Chief Marketing Officer at Cognigy, and Charles Pointer, Vice President of Bank Operations at Commerce Bank. During their respective episodes, both leaders discussed in great detail how to effectively implement AI in customer service by striking a balance between automation and human support and integrating it seamlessly into existing systems.
Their conversations underscore the importance of solid foundations, such as clean data, system integration, and cross-team alignment, to ensure AI delivers tangible value at scale. This article examines two key insights from their conversations for financial services leaders adopting new and advanced forms of AI at their organizations:
Prioritizing integration and governance for agentic AI success: Integrating legacy systems, selecting the right partner, and adopting a phased approach to ensure effective governance and success for new AI-driven agentic systems.
Improving knowledge systems to enable accurate AI responses: Upgrading documentation and SOPs to ensure AI pulls current, reliable information when assisting agents or customers.
Prioritizing Integration and Governance for Agentic AI Success
Episode: Rethinking Customer Experiences with AI-Driven Conversations – with Alan Ranger of Cognigy
Guest: Alan Ranger, Chief Marketing Officer at Cognigy
Expertise: Business Development, Strategic Partnership, Marketing
Brief Recognition: Before joining Cognigy as Vice President of Marketing, he led global market development at LivePerson, where he spent six years driving international growth. Earlier in his 30-year career, he held various sales, marketing, and leadership roles across both startups and large enterprise software companies.
Alan explains that while it’s easy to build an AI chatbot that can hold a conversation, it’s much harder to get it to complete fundamental tasks. For that, AI needs to be connected to all the back-end systems a company uses.
Alan cites an example of a large insurance company where all customer calls — tens of millions annually — are first answered by AI. The system:
Identifies the caller
Verifies their identity
Understands the reason for the call
Retrieves relevant information from legacy systems
Passes the full context to a human agent for resolution
For instance, if someone calls after a car accident to check if they can get a rental, the AI fetches the details and then hands the call off to a human agent, who is now already equipped with the full context.
“But not only does it do the handover, it changes role and becomes a copilot, and it’s already got the context of the whole conversation, so it knows exactly what’s been said, what the issue is, and it can then do the next best action, prompts, all that sort of thing.
Again, in a compliance environment, it can make sure that certain statements are read. And then at the end of the call, it does the wrap-up. The human agent checks it, and then the AI agent does all of the updates in those old legacy back-end systems.”
—Alan Ranger, Chief Marketing Officer at Cognigy
Alan says the first step for any company looking to implement AI agents is to choose the right partner. He explains that many companies in the market are simply wrapping an LLM in a wrapper and calling it an AI agent. These tools may look impressive in demos, but they often lack the in-depth enterprise knowledge and scalability required in real-world situations.
He gives the example of a sudden surge in calls, such as if an airport shuts down and 10,000 calls come in at once. Most off-the-shelf tools won’t be able to handle that kind of spike. A good partner, he says, needs to understand enterprise-scale and surge capacity, which is something that can’t be solved by hiring more people.
Beyond scale, Alan highlights the importance of orchestration. That means being able to integrate with legacy systems. He points out that most large enterprises lack modern tech stacks or a single unified platform. Instead, they have different systems, and there’s no plug-and-play AI agent that works with all of them.
The build part addresses that challenge. Alan says a strong platform should come with prebuilt integrations that help shorten time to value. While no system will work seamlessly with every legacy tool out of the box, having a platform designed for enterprise integration can significantly accelerate and streamline the process.
Such a partner, he advises, is exactly what businesses need.
Alan says that for risk-averse companies, a smart first step is choosing a high-volume but simple use case suitable for traditional, rule-based conversational AI. It helps them build internal knowledge, understand integration, and set the stage for more advanced AI.
His key advice is to form an AI Council. These are cross-functional teams that align on ethics, compliance, vendors, and strategy. It keeps the business focused, avoids one-off distractions from flashy demos, and helps get AI solutions into production faster and with fewer roadblocks.
Improving Knowledge Systems to Enable Accurate AI Responses
Episode: Why Human-Centered Service Still Matters in the Age of AI – Charles Pointer at Commerce Bank
Guest: Charles Pointer, Vice President of Bank Operations, Commerce Bank
Expertise: Banking Operations, Operations Management, Customer Experience
Brief Recognition: Charles Pointer currently leads bank operations at Commerce Bank, bringing over a decade of experience in retail banking and customer experience. Before his current role, he held leadership positions at UMB Bank and Bank of America, with a focus on branch management and operational strategy. He earned his MBA from Rockhurst University. Previous to his experience at Commerce Bank, Charles led call center operations as Director for Waddell & Reed.
In his podcast, Charles shares that by studying call types and utilizing speech analytics, his team aims to strike a balance between self-service and personalized support, thereby enhancing both employee and customer experiences.
Charles emphasizes the importance of aligning different teams across the bank, each with its own goals, expectations, and approach to customer support. He’s focused on understanding what the bank promises customers and how internal teams can collaborate to resolve common issues.
“I always want to understand, what are we telling our customers? What are we selling to our customers, and how are we talking internally with, say, our product team to say we’re getting a lot of password reset questions. How can we eliminate that by enhancing our product and our service to automatically be able to push self service more as part of the product itself.
How do we see people go from phone calls to emails? How can we just eliminate all of those inquiries coming in where we can stop it on the front end versus getting it on the back end.”
– Charles Pointer, Vice President of Bank Operations at Commerce Bank
He’s concerned that outdated knowledge base content could cause AI to give incorrect answers. To prevent this, he advocates improving knowledge management systems to ensure accurate, up-to-date documentation is available for agents – both human and digital – to reference.
Charles also sees opportunities in tools such as speech analytics and agentic AI that can automate tasks such as call scoring and quality checks that are currently performed manually. These may not be complete AI solutions, but they could still drive immediate impact.
He emphasizes the importance of planning the tech stack in the correct order. Moving too fast without foundational systems in place could lead to costly missteps. To avoid this, he’s considering bringing in outside experts to guide implementation and ensure each step builds effectively on the last.
Ultimately, he emphasizes the importance of selecting vendor tools that can integrate with existing systems, such as the CRM, rather than replacing them. He values solutions that enhance current platforms by enriching the data and improving service without forcing significant system changes.