Customer Chatbots: The Good, The Bad, and the Ugly

Chatbots have changed how businesses interact with customers - sometimes for the better, sometimes not so much.

They offer quick responses, scalability, and the convenience of round-the-clock availability.

But, like any tool, their effectiveness depends heavily on how they’re implemented and used.

Some enhance the customer experience beautifully, while others leave users feeling frustrated and unsupported.

Let’s explore both the high points and pitfalls of chatbots in customer service.

The Good

At their best, chatbots bring huge advantages to companies - especially in handling everyday customer needs quickly and efficiently.

They help businesses save time, cut costs, and deliver service at speeds humans simply can’t match.

24/7 Availability

One major selling point? Chatbots never sleep.

Whether it’s 2 PM or 2 AM, they stand ready to assist customers, without the need for breaks, shifts, or holidays.

And it’s paying off - 64% of users say 24-hour access is their favourite thing about chatbot support.

"Ensuring customer communication remains secure and protected, even when handled by chatbots, is critical in today’s digital landscape. Trust is everything."

Paul Holland, CEO, Beyond Encryption

Efficient Management of Routine Tasks

For repetitive, predictable queries - like checking order status or resetting passwords - chatbots are in their element.

Machine learning models even allow some bots to improve over time, offering faster and more accurate responses with each interaction.

It’s estimated that chatbots can handle up to 80% of routine customer service questions - freeing up human agents for more complex work.

Cost-Effective Solution

By taking on the heavy lifting of everyday support, chatbots can drive serious savings.

Some studies suggest businesses can reduce customer support costs by up to 30% simply by using bots strategically.

"The integration of security protocols within AI interactions is vital for customer trust and regulatory compliance."

Mike Wakefield, CTO, Beyond Encryption

Quick Response Times

No one likes waiting in a queue. Chatbots offer almost-instant answers to common queries, keeping customers happy and conversations moving.

In fact, 62% of consumers say they’d prefer an immediate chatbot conversation to waiting 15 minutes or more for a live agent.

Data Collection and Analysis

Chatbots don’t just talk - they listen, too.

Every customer interaction provides data points that can help businesses better understand their audience’s preferences, pain points, and needs.

According to research, 55% of businesses using chatbots generate more high-quality leads.

"Understanding customer behaviour through AI-driven data analysis provides businesses with an opportunity to personalise services and create meaningful experiences."

Adam Byford, CCO, Beyond Encryption

The Bad

Of course, not everything is rosy when it comes to chatbots.

Despite all their strengths, they still have real limitations - and when things go wrong, customers can get frustrated fast.

Limited Understanding

While chatbots are pros at answering straightforward questions, more complicated or emotionally sensitive queries? That’s where they often stumble.

Many still rely on pre-set scripts or narrow training sets, leading to answers that can feel robotic - or miss the mark entirely.

No surprise then that 59% of customers feel that chatbots often misunderstand the nuances of human communication.

Lack of Empathy

No matter how "smart" an AI gets, it still can't replicate a human's ability to truly empathise.

When emotions run high - complaints, cancellations, sensitive issues - customers want someone who can genuinely listen, not just spit out canned responses.

60% of consumers believe humans simply understand their needs better than any bot ever could.

Technical Glitches

Then there’s the unavoidable: bugs, errors, and crashes.

Chatbots - like any software - are vulnerable to malfunctions that can leave customers stranded mid-conversation.

According to a survey by Userlike, 43% of customers reported experiencing technical problems when interacting with bots.

Language and Cultural Barriers

Even with today’s advances in natural language processing, chatbots still sometimes struggle across languages and cultural nuances.

This is a real hurdle for global companies: only 29% of businesses have successfully rolled out multilingual bots.

The Ugly

Beyond day-to-day limitations, there are deeper - and sometimes riskier - concerns about chatbot technology that businesses can’t afford to ignore.

Privacy and Security Risks

When chatbots collect personal data, they become a juicy target for cybercriminals.

Weak security protocols can lead to breaches, exposing sensitive customer information and wrecking brand trust - not to mention causing regulatory headaches.

The NCSC has warned specifically about the risks linked to large language models (LLMs) and AI-driven chatbots.

"Privacy and security must remain a priority in any AI communication. Customers need to know that their data is safe, regardless of how they're engaging with a company."

Emily Plummer, Marketing Director, Beyond Encryption

Misinformation and Liability Issues

Sometimes, a chatbot doesn’t just give the wrong answer - it gives advice that creates legal trouble.

Remember when Air Canada's chatbot mistakenly misinformed a customer about bereavement fares, leading to a lawsuit? That case made headlines for all the wrong reasons.

Ethical Concerns

As bots get smarter, transparency gets murkier.

Are customers always aware they’re speaking to a bot? Should they be?

AI systems can unintentionally reinforce biases if their training data is skewed - a potential reputational and ethical landmine for companies.

Job Displacement

Automation doesn’t just make processes faster - it also threatens jobs.

The World Economic Forum predicts that automation will disrupt 85 million jobs globally by 2025, reshaping the customer service landscape in ways we’re still trying to fully understand.

Potential Alternatives

Given the strengths and weaknesses of chatbots, a smart strategy isn’t about replacing humans - it’s about combining the best of both worlds.

Here are a few practical alternatives (or complements) to chatbot-only service models:

Human Agents

Deploying a hybrid model - where chatbots handle basic queries and escalate complex ones to human agents - offers a safety net for tricky conversations that bots can’t navigate.

It keeps operations efficient while ensuring customers who need a real person can easily connect with one.

Self-Service Portals

Sometimes, customers don’t need to talk to anyone - bot or human. They just want answers.

Robust self-service hubs with FAQs, troubleshooting guides, and community forums empower users to solve issues quickly, without waiting for support at all.

Video Calls

For products that are visual or need a personal touch, video support can be a game-changer.

Retailer DFS, for example, saw customer satisfaction soar above 90% by offering virtual shopping consultations - outperforming traditional phone service significantly.

Secure Email

For sensitive conversations - billing issues, personal data exchanges, legal disputes - secure email is a safer, more reassuring option than real-time chatbots.

Services with encryption and verified delivery provide peace of mind for customers who prioritise security just as much as convenience.

Striking the Right Balance

Chatbots are here to stay - but they’re not a silver bullet.

The best customer experiences come from a careful blend of technology and human support.

When businesses let chatbots handle the quick, simple stuff - and leave more complex or emotional interactions to trained humans - everybody wins.

Secure channels like encrypted email and video calls round out the approach, ensuring that every customer interaction feels personal, professional, and protected.

Done thoughtfully, businesses can create a customer journey that’s not only fast and efficient but also empathetic and trustworthy.

References

18 Customer Service Chatbot Examples, HubSpot, 2023

Cut Costs, Boost Service: The 24/7 Power of AI Chatbots, Potential, 2024

Customer Service Chatbots: Everything You Need to Know, Talkative, 2022

Survey: Customer Expectations for Chatbots Are..., Chain Store Age, 2022

Chatbots: Friend or Foe? How They’re Reshaping Customer Experience, Gabba, 2024

86 Critical Chatbot Statistics: 2021 Data Analysis & Market Share, Finances Online, 2021

How AI Chatbots Can Break Down Language Barriers, FastBots, 2024

Air Canada Chatbot Case Highlights AI Liability Risks, Pinsent Masons, 2024

Recession and Automation Changes Our Future of Work, World Economic Forum, 2020

Reviewed by

Sabrina McClune, 13.09.24

Sam Kendall, 19.09.24