CX automation has a trust problem. Our survey shows customers still want to give it another chance.

Every customer has a support horror story to tell. The chatbot that endlessly looped them through the same three options. The voice system that couldn’t understand them. The hold music that played for 40 minutes before the call dropped. They share these stories with anyone who will listen, including your prospective customers.
People tell these stories not just out of frustration but as a way to warn the next victim of a poor support experience. The failed support cases, both experienced and shared, cause a general distrust in automated support that encapsulates most consumers.
Parloa’s Consumer Patience Index 2026 shows just how deep that trust gap runs. By surveying 1,001 US consumers who recently interacted with customer service systems, we found that while customers are deeply frustrated with today’s automation, they’re still willing to give it another chance if someone builds it better.
Customers also told us exactly what winning back their trust with automated support looks like. Their feedback offers a practical roadmap for the brands that want to build automation customers will actually use, trust, and return to.
Disdain for deflection
When we asked customers to rank their most frustrating support experiences, talking to a bot that doesn't understand them ranked number one. And they have very little patience for waiting for the system to understand. 60.1% of respondents said they’d only tolerate repeating themselves twice before giving up entirely.
This is the predictable result of automation designed to deflect volume rather than resolve problems. Customers have learned to expect failure: More than half (53.6%) admit to actively gaming automated systems just to reach a human faster.
Data validates that this frustration is a feature, not a bug. Earlier this year, Parloa's State of Agentic CX study assessed the support options of more than 10,000 enterprise websites and found that only 8.9% of chatbot conversations achieved the customer's stated goal.
Before customers can trust automation again, they need proof that it can understand them.
Loyalty to improvement
The survey's data points to a real opportunity:
84.9% of customers say they'd keep using automated systems if those systems resolved their issues consistently.
Nearly 70% believe future AI will handle complex service requests better than humans.
75.2% say they'd prefer automated support if it could anticipate their needs proactively.
Customers aren't avoiding automation, they’re just waiting for proof that it can actually work. Brands that close the gap between current performance and customer expectations will have a clear advantage.
What "working automation" actually means
The blueprint customers outlined comes down to three key elements: comprehension, speed, and continuity.
Comprehension
People just want to feel understood, even when there’s no human on the other end. They expect automation to grasp their intent, remember context, and respond appropriately. If it can’t do that, trust disappears quickly.
Speed
Customers don’t expect every issue to be resolved instantly, but they do expect momentum. More than half (55.5%) of respondents said they’d disengage within three minutes if resolution wasn’t happening.
Continuity
Not every issue can be solved through automation alone. When a human agent needs to step in, customers expect their context to come with them. Every time customers have to repeat themselves, trust erodes.
This repetition has led to a common assumption: 85.4% of customers reported being forgiving to the human agent when repeating themselves after a routing because they had assumed the automated system would fail from the start. The brands that can upend that expectation are the ones customers will trust.
Give your customers a better story to tell
The bar is clear. Brands that earn trust through better automation will be the ones customers can't stop talking about, for all the right reasons.
Read the Parloa Consumer Patience Index report to understand more of what's driving consumer frustration today, and what they're hoping to see from the next generation of automation.
:format(webp))
:format(webp))
:format(webp))
:format(webp))