Meanwhile, friction before, during, and after the calls continue to slow agents down and frustrate customers. As discussed below, repetitive authentication, hard-to-navigate systems, scattered knowledge, unclear dashboards, and staffing models that prioritize coverage over capability are just a few examples of it.
The result? Calls may move faster on paper, but they rarely move better.
In this three-part article, I am exploring how contact centers can reduce handle time without rushing the customer or the agent: by redesigning authentication, improving knowledge access, rethinking staffing and specialization, and strengthening agent support, to name a few.
Part 1 covers removing customer friction before the call, Part 2 will focus on agent performance strategies and methods, and Part 3 will discuss having the right people in the right place and the tools to help them; it will also present the conclusion.
For when done right, efficiency and CX don ' t compete. Instead, they should reinforce each other.
THE REAL PROBLEM ISN ' T TALK TIME: IT ' S FRICTION
When leaders look at handle time, their instinct is often to focus on what happens during the conversations:
• How long the agents talk.
• How many pauses there are.
• How quickly agents sit in after call work( ACW).
But in most centers, talk time is not the primary driver of high AHT. Friction is.
Agents regularly lose minutes- not seconds- to accomplishing tasks that have nothing to do with serving the customers:
• Authenticating callers multiple times and / or using long and drawn-out authentication processes.
WHEN FRICTION DOMINATES THE INTERACTIONS, NO AMOUNT OF COACHING ON SPEED WILL CRE- ATE MEANINGFUL IMPROVEMENTS.
• Navigating disconnected systems, searching for information that should be easy to find.
• Putting customers on hold to confirm answers they should already have or systems taking too long to navigate between screens.
As many agents put it in online forums, " The customer isn ' t slowing me down. The systems are."
COACH’ S CORNER
From what I’ ve seen, when call volumes rise and pressure mounts, the response is rarely systemic. Instead of fixing what’ s slowing agents down, management tightens the levers closest to them- asking agents to talk less, hold less, and wrap up faster- without addressing why those delays exist in the first place.
When friction dominates the interactions, no amount of coaching on speed will create meaningful improvements. In fact, it often does the opposite: agents feel pressured, customers feel rushed, and repeat calls increase, thereby reducing first call resolution( FCR). If organizations want to reduce handle time without damaging the experience, the first step is to stop treating AHT as an agent behavior problem and start treating it as a design problem.
REMOVE FRICTION BEFORE THE AGENT SAYS HELLO
One of the most effective- and underutilized- ways to reduce handle time is to redesign the steps that occur before the call reaches an agent.
PACING AGENT CROSS-TRAINING
Cross-training is often promoted as a quick way to improve the average speed of answer( ASA) and increase scheduling flexibility.
When done thoughtfully, this method can be effective. But when it becomes the default solution- especially for complex call types- it often leads to the opposite: longer handle times, lower FCR, and unnecessary transfers.
It is important to remember that not all queues are equal. Some require weeks of training due to regulatory requirements, complex workflows, or nuanced customer needs. In these cases, efficiency doesn ' t come from breadth: it comes from depth.
A more sustainable approach is to build expertise first. New hires can begin in lower-complexity queues, followed by a structured nesting period with mentoring and coaching. Once performance stabilizes- reflected in quality scores, confidence, and consistency- additional queues can be introduced gradually.
When cross-training is paced correctly, centers see real benefits:
• Increased call variety without overwhelming agents.
• Clear development paths that support employee engagement.
• Improved occupancy without sacrificing resolution quality.
Problems arise when multiple complex queues are added too quickly. As a seasoned agent put it, being trained on everything can feel like being an expert at nothing.
Broader cross-training can still make sense for simpler, transactional queues with short learning curves. Even then, skill expansion should be driven by data: not coverage pressure.
Reviewing quality assurance( QA) scores, performance trends, and agent readiness before adding new skills helps prevent unintended increases in handle time.
MAY 2026 11