Should customer service agents be promoting self-service?

CX Insights

Should agents be promoting self-service?

Gartner found 60% of customer service agents “fail to promote self-service” to customers.

This study struck me as interesting because 81% of people want self-service and try to do so before reaching out to support. With the rise of chatbots and even AI agents, if a customer connects to a human, it’s very likely because they need or want the help from a human agent.

Plus, I remember in my days as a frontline agent, customers could be quite sensitive if we pointed out things they could do for themselves. Some might perceive it as rude or being unwilling to help, and I wasn’t willing to jeopardize my CSAT score (which factored into my compensation) over it!

What’s a better solution to encouraging customers to self-serve? Making those resources accessible to customers - not loading yet another thing onto your agents who are focused on solving problems for customers.

Get our guide on how to improve your brand’s self-service for all the tactics.

CX: Coming to a mass transit agency near you

Nearly 70 U.S. and Canadian transit agencies are investing in customer experience (CX) to improve ridership, satisfaction, loyalty, and safety. Programs span CX departments, journey mapping, real-time feedback, and employee training.

I’ve lived in three places spanning the full spectrum of quality of public transportation in the US (including NYC-ish right now) but just got back from my first trip to London and was blown away by the ease of use of the transit system. Infrastructure is hard to replicate, but clear signage, overhead announcements, great lighting, human staffing at major stations, and easy payment are all other ways to make the rider experience better, and much quicker wins.

No matter the industry you’re in, the key takeaway is how thinking cross-functionally and focusing on user-centric design can boost adoption and advocacy.

Building trust with call recording consent

When your customers trust you, they give you their loyalty. But trust doesn’t just happen; it’s built through every interaction they have with your business. Mishandling customer data or failing to comply with regulations can quickly erode that trust.

At Kustomer, we’re committed to helping businesses deliver exceptional service while adhering to the highest standards of compliance. Read more on the blog about our new Automated Call Recording Consent feature.

CX Quotation of the Week

“You don't automate things completely that you care about. You automate the things you don't care about. You won't automate the relationship with your parents, your partner, your kids. When you say, ‘I wanna just automate quality’, you might as well say, ‘I don't care about quality.’ What the best are doing is using humans to pay attention to quality, but they're using AI to pay attention to other signals outside of CSAT or AHT.”

-Vasu Prathipati, CEO and co-founder, MaestroQA

AI News

Longer memory key in AI-powered CX

New multimodal and memory-enhanced AI models from Meta, OpenAI, and others are transforming customer experience. These tools can process voice, images, and text — and remember customer interactions over time — enabling hyper-personalized, human-like service at scale. For CX leaders, the opportunity lies in using AI to deliver seamless, context-rich experiences while maintaining trust through strong data privacy practices.

Notably the timeframe for the conversations was December 2020 through November 2021 - and AI-powered customer support technology has come a long way since then. With true AI agents, these numbers will likely improve further.

Claude 4’s 7-hour shift: A milestone for long-running AI

Anthropic’s Claude 4 can now work on tasks for seven straight hours — a major leap in AI endurance and memory. For CX leaders, this unlocks potential for more sustained, context-rich customer interactions. However, 24/7 AI remains limited by high computing costs, memory constraints, and sustainability concerns.

The future may require balancing performance with ethics and environmental impact. This week, Meta signed a 20-year deal to buy nuclear power (starting in mid-2027) to meet power demand.

Banks feel like they are falling behind in AI adoption

Insights are coming out of the American Banker Digital Banking Conference. In short: banks feel like they are falling behind in AI adoption due to the rise of agentic AI. Here are some findings:

52% of banks have fully implemented chatbots
Less than 35% of banks have fully implemented almost any other AI-powered solution (such as virtual assistants, document processing, predictive analytics, anti-money laundering)
The most common challenge to banks adopting AI is legacy infrastructure that can’t support new processes

Anyone who has worked at a bank (me!) knows they move glacially in part due to regulations, but also deeply entrenched culture. That’s part of why fintechs are gaining so much market share.

CX leaders in the financial services realm: this is an area that’s worth pushing to be ahead of the pack. Explore AI-powered customer support solutions that meet customer needs; if you don’t, someone else will.