Why call deflection is no longer the contact center's north star
Leaders should evaluate automation and self-service based on whether they resolve customer needs efficiently and durably, not simply whether they prevent a call or human handoff.
Historically, contact center leaders have been obsessed with measuring call deflection -- that is, how many customer service calls they can avoid referring to human agents. By redirecting incoming calls to alternative self-service or digital support channels, companies aim to reduce queue volumes and resolve customer issues faster. When done right, deflection can also improve agent productivity and reduce operational costs. But despite its potential benefits, this one metric alone is not sufficient as a primary measure of contact center performance.
Rather than focus solely on call deflection, contact center leaders should aim to balance self-service efficiency with resolution quality, customer effort and reduced repeat contacts.
Where call deflection works -- and where it doesn't
Call deflection involves redirecting a portion of customer service calls to digital or self-service channels, such as automated chatbots, knowledge bases and AI agents. Through such automations, customers can complete a wide range of simple tasks without human intervention. The customer is happy that their issue is resolved without a significant wait time, agents are happy because their workload is lower, and the company is happy that it can alleviate high call volumes and save money.
In a recent survey conducted by Salesforce, 61% of participants said they would prefer self-service to resolve simple issues. That said, call deflection is not always a win-win situation. While it is a convenient and valuable approach for simple, repetitive, high-volume and transactional requests, automated digital channels cannot resolve every complaint -- particularly complex ones. And a failure to resolve issues is the No. 1 frustration when using chatbots for 43% of survey respondents, according to a 2025 Zoom-commissioned survey by Morning Consult. "Getting stuck in a loop" is another cause for friction, further contributing to resolution failure and driving increased brand abandonment and customer churn.
These systems sometimes struggle to seamlessly transfer the full context of automated conversations to human agents, creating operational bottlenecks that slow down issue resolution and even damage CX. The same Zoom-commissioned survey cited above found that human escalation is a "critical" failure point for customer service, with a huge gap existing between what customers expect -- i.e., human escalation when needed -- and what actually happens --i.e., little or no escalation. Poor CX has a significant long-term effect on customer loyalty, with 60% of customers admitting that even one or two poor interactions are enough to cause them to leave the brand.
The resolution economy
In a February 2026 article, Zoom introduced the idea of the "resolution economy." This term refers to the idea that the speed at which companies resolve customer issues does not necessarily indicate a high-performing customer service department. Instead, Zoom argued that other factors, such as FCR and reduced repeat contacts, play a bigger role in improving a firm's customer service performance, reducing its operational load and strengthening its competitive advantage.
Why call deflection is a misleading metric
For contact centers everywhere, a major priority is minimizing the cost of customer contact. And for years, call deflection was the preferred approach to meet this goal. As it remained the north star for contact centers, investment into virtually every automated chatbot, interactive voice response (IVR) menu, searchable knowledge base and self-service portal was justified by its potential ability to reduce call volumes and deliver a clear operational advantage.
Deflecting calls to an automated system can speed up calls, accelerate resolution and lower costs. However, as a success metric, call deflection is often disconnected from more meaningful outcomes such as customer retention, customer satisfaction (CSAT) and revenue, which can better clarify an organization's performance and guide its strategy.
Here are two examples where call deflection can be a misleading metric.
High deflection, low CSAT
A retail firm's contact center platform automatically deflects 100 customers per day to a chatbot. The system's basic menus and standard scripts are sufficient for simple tasks such as tracking package delivery and initiating returns. However, for more complex issues such as suspected credit card fraud or damaged-item complaints, callers might simply abandon the automated system without finding a solution to the problem. Thus, in this scenario, a low CSAT score is likely to accompany a high call deflection rate.
High deflection, high customer effort
A B2B SaaS company deflects 60 customers per week to its knowledge base of product documentation, video tutorials and troubleshooting guides. Although useful for everyday tasks such as account creation and product debugging, these resources cannot help the organization's enterprise customers with critical issues such as platform outages or contract changes. To resolve these issues, customers need to speak with human agents, which increases customer effort scores. In this way, a supposedly deflected interaction could end up causing repeated contacts through other channels.
Looking beyond the narrow confines of call deflection
While automating some customer interactions can reduce contact center costs and, in many cases, speed up response times, it is important not to over-rely on call deflection to gain a full picture of contact center performance.
It is also critical for contact center leaders to ask the following questions:
- Did automation help more customers find the answers they were looking for, or did it simply force them into unhelpful loops or transitions between channels?
- Did limiting agent contact lower handling times while leaving the underlying issue unresolved?
- Are customers spending more time and effort on IVR or self-service instead of speaking with a human agent?
- Does the system appear to be successful based on containment or deflection rates, even though CSAT scores are low?
- Does the high call deflection rate directly correlate with a high FCR rate?
Gaining an understanding of these questions can provide leaders with greater clarity on how the contact center is performing, how well it aligns with customer expectations and to what extent it delivers real business value.
From deflection to resolution: The way ahead for contact center leaders
For modern contact center leaders, the primary objective needs to shift from reducing human-to-human interactions to resolving issues at first contact. In other words, they need to design automations and self-service around successful resolution rather than maximum containment.
Moreover, the focus should be on FCR across the entire customer journey, not just within one channel. This is essential to ensure that customers do not face additional obstacles or waste time while seeking to resolve a problem.
A clear escalation path is also vital. Users should be able to access a human expert when automation is unlikely to resolve the issue. A built-in escalation workflow will ensure that the automated system seamlessly hands off a complex interaction to a human agent with full context. This approach signals respect for the customer and can help the organization transition from impersonal transactional interactions to more personalized customer conversations.
Journey data and AI-enabled analytics tools can help organizations identify complexity during the customer journey, surfacing issues like customer frustration and repeated contact attempts. They can also track how customers engage with the brand across multiple channels over time and take concrete action to alleviate points of friction.
Leaders should also aim for a more balanced contact center scorecard. This scorecard considers several important metrics:
- Issue resolution quality and durability.
- Time to complete resolution, including transfers and follow-up contacts.
- Total cost of resolving the issue rather than simply the cost of the first interaction.
- Customer effort and satisfaction following a service interaction.
- Repeat contacts for the same or related issue.
Finally, leaders should connect contact center performance to customer retention, churn, revenue and customer lifetime value (CLV) rather than just call deflection rates. It is equally important to redesign agent incentives so they are driven primarily by FCR, CSAT, Net Promoter Score and internal QA scores.
A balanced approach is key to contact center success
In a customer-driven business landscape, strong customer relationships are often the key differentiator between successful organizations and those that struggle. The latter focus only on fast issue resolution and obsess over call deflection rates. At the other end of the spectrum are the companies that treat deflection exactly as what it is: a useful metric that still matters for simple, transactional issues but is not the best way to measure or optimize contact center performance. These organizations are not looking to eliminate self-service but to make it genuinely helpful to customers. And this is how they develop and maintain strong customer relationships.
By shifting the focus from "How many calls can we deflect?" to "How well do we solve customer problems?" successful contact centers put the customer squarely at the center of every conversation. Only by shifting from a cost center mindset to a relationship investment mindset and by prioritizing resolution metrics over deflection metrics can they win customer loyalty and boost CLV.
Rahul Awati is a PMP-certified project manager with IT infrastructure experience spanning storage, compute and enterprise networking.
