Customers rarely contact an organisation because they want to interact with its contact centre. They want an answer, a problem resolved or something completed. Sometimes that means finding information on a website. Sometimes it requires a short conversation through live chat. Increasingly, it might involve an AI-powered virtual agent capable of answering questions or completing simple transactions.
But when self-service fails, customers can find themselves navigating a chatbot, repeating information and eventually contacting a human agent anyway.
The technology may have reduced the number of calls arriving directly at the contact centre. It has not necessarily reduced customer effort.
That distinction should shape how organisations evaluate web self-service and chat solutions.
The objective isn’t to prevent customers from reaching an agent. It’s to help them reach a satisfactory resolution with as little unnecessary effort as possible.
Start with the Customer’s Reason for Contact
Before selecting a chatbot or live chat platform, contact centre teams should understand why customers are getting in touch.
Which enquiries occur most frequently? Which are relatively straightforward? Which require access to account information or back-office systems? Which involve complex decisions, vulnerability or circumstances where human judgement matters?
A customer checking delivery information has different requirements from somebody disputing a financial transaction or seeking support during a distressing situation.
These differences should determine which interactions are suitable for automation and which should remain with human advisers.
Ciptex, for example, has worked with housing charity Shelter on a multichannel contact centre incorporating live chat and chatbot-based triage, with the aim of helping people access information while allowing advisers to focus on more urgent cases.
That illustrates an important principle: the most valuable use of self-service is not always eliminating the interaction. It may be identifying the customer’s need and directing them towards the most appropriate support.
Understand the Different Self-Service Options
Web self-service covers several distinct capabilities.
A searchable knowledge base can help customers find answers independently. Guided journeys can take users through a structured series of questions. Traditional chatbots may use predefined rules and responses, while conversational AI can interpret less structured requests.
More advanced AI agents may also interact with connected systems to complete approved actions, such as updating account information or checking an order.
Live chat serves a different purpose by giving customers access to a human adviser through a digital channel.
Capacity provides AI-enabled customer support, knowledge management and workflow automation, while Zoom offers Virtual Agent technology alongside its broader Contact Center platform.
The important comparison is not which provider offers the most sophisticated chatbot.
It is which combination of knowledge, automation and human support can resolve the organisation’s actual customer enquiries.
Build Self-Service Around Trusted Knowledge
A chatbot cannot reliably answer questions if the information it uses is incomplete, inconsistent or out of date.
That makes knowledge management a foundation of effective self-service.
Contact centre teams should establish which information sources are authoritative, how content is updated and who owns its accuracy.
Capacity, for example, connects AI-powered support with knowledge-base information and helpdesk workflows, helping organisations use existing knowledge to answer enquiries and support agents.
But simply connecting a generative AI model to a collection of documents is not enough.
Organisations should test whether the system retrieves the correct information, distinguishes current policies from obsolete versions and recognises when it cannot provide a reliable answer.
A confident but incorrect response can create more work than an unanswered question.
The strongest self-service systems should know when to provide an answer and when to stop.
Compare Resolution, Not Just Deflection
Contact centres frequently measure self-service through containment or deflection rates.
These can be useful indicators of how many interactions are handled without a human agent.
However, an interaction ending inside a chatbot does not necessarily mean the customer’s problem was resolved.
The customer might abandon the conversation, search elsewhere or contact the organisation again.
Buyers should therefore examine how providers measure successful outcomes.
Can the platform distinguish completed transactions from abandoned conversations? Can it identify repeat contacts? Does it collect meaningful customer feedback? Can managers examine conversations where the system failed to understand the request?
The question should be: Did the customer get what they needed, or did the interaction simply end?
That distinction becomes especially important as businesses automate a larger proportion of digital enquiries.
Make Human Escalation Easy
Not every interaction should be automated from beginning to end.
Some customers will need human assistance because the enquiry is complex, sensitive or unusual. Others may simply prefer to speak to somebody.
The transition between self-service and human support is therefore one of the most important capabilities to evaluate.
HGS combines conversational AI with customer-service operations, including approaches designed to transfer complex interactions to human advisers while preserving relevant context.
Kerv provides contact centre technology and AI chatbot services, including integrations with platforms such as Genesys Cloud.
Buyers should establish whether customers can request an adviser, how escalation decisions are made and whether the human agent receives a usable summary of what has already happened.
A good handover should feel like the same conversation continuing, not a new conversation beginning.
Connect Chat with the Wider Contact Centre
A website chatbot may work effectively in isolation while creating problems elsewhere.
Customers often move between channels. They might begin with a chatbot, continue through live chat and later make a telephone call.
If the systems involved cannot share context, the customer may need to explain the same issue repeatedly.
Gnatta provides digital customer engagement and contact centre technology relevant to managing conversations across channels, while Wavenet supports unified contact centre environments bringing together communications, automation and customer information.
Zoom Contact Center also supports web chat alongside other contact channels, with Zoom Virtual Agent providing automated conversational support.
For buyers, integration should be assessed at the customer-journey level rather than simply through a list of supported channels.
Can agents see previous interactions? Can the system identify an existing case? Does the customer retain their place in the process when moving between channels?
True omnichannel service should reduce repetition.
Look Beyond Answers to Actions
Providing information is valuable, but many customers contact organisations because they want something done.
They may need to change an appointment, update their details, report a problem or request a refund.
Where appropriate, connecting self-service technology to CRM, order-management and other business systems can allow these activities to be completed without human intervention.
Wavenet offers AI-enabled self-service and automation capabilities, while Capacity combines customer-support knowledge with workflow automation.
IP Integration provides contact centre solutions and integration capabilities relevant to connecting customer interactions with wider communications and business processes.
However, allowing automated systems to take action creates additional requirements.
Organisations need to determine which actions can safely be automated, what information is needed to authorise them and when human review is necessary.
Answering a question and changing a customer’s account are different levels of responsibility.
Don’t Overlook Live Chat Operations
Live chat is sometimes treated as a relatively simple extension of digital customer service.
But operating it effectively requires staffing, routing, supervision and performance management.
Agents may handle more than one chat simultaneously, depending on the complexity of the interactions and the support available.
Excessive concurrency can make responses slower, reduce quality and increase the likelihood of mistakes.
Businesses also need to decide when live chat will be available, what happens outside operating hours and whether different enquiries require specialist teams.
Magellan Solutions offers outsourced live chat support alongside wider customer-service operations, providing an alternative for organisations that do not want to deliver every digital interaction internally.
ellio, meanwhile, combines customer experience operations, technology and advisory services, including support for organisations operating in complex or regulated environments.
For buyers, the decision may involve technology, staffing or a combination of both.
Design for Accessibility and Customer Choice
Self-service should make support easier to access, not introduce a new barrier.
Some customers may have disabilities, limited digital confidence, language requirements or circumstances that make automated conversations unsuitable.
The UK’s public-sector accessibility requirements and the wider Web Content Accessibility Guidelines provide useful benchmarks for accessible digital services.
Contact centre teams should assess whether web chat works with assistive technologies, supports keyboard navigation and presents information clearly.
They should also consider whether customers can access an alternative support channel when needed.
Customer choice matters because the quickest interaction for one person may be frustrating or inaccessible to another.
Address Privacy and Security Early
Self-service and live chat platforms may process names, contact information, account details and other personal data.
AI-enabled systems can introduce additional considerations around how information is accessed, processed and retained.
The ICO’s guidance on AI and data protection emphasises the need to protect individuals’ rights when personal data is used within AI systems, including systems supplied by third parties.
Organisations should understand what information the platform collects, where it is processed, who can access it and whether it may be used for purposes beyond delivering the customer interaction.
Authentication and authorisation also matter when a chatbot can retrieve account information or perform transactions.
Customers should be able to ask general questions without unnecessary identification, while more sensitive activities may require stronger verification.
Security should follow the risk associated with the interaction.
Test AI Against Real Customer Conversations
AI demonstrations can make difficult customer enquiries appear straightforward.
Real customer language is considerably less predictable.
People use different terminology, provide incomplete information, make spelling mistakes, change their minds and ask several questions at once.
Before deployment, organisations should test systems against realistic customer enquiries, including ambiguous requests and situations where the correct response is to escalate.
Testing should also examine what happens when systems are unavailable or information cannot be retrieved.
Zoom Virtual Agent, Capacity and other conversational AI platforms provide capabilities for automated customer interactions, but the quality of the experience depends heavily on configuration, knowledge and the workflows surrounding the technology.
A fluent answer is not necessarily an accurate or useful one.
Measure Customer Effort and Operational Value
Self-service can reduce the workload associated with routine enquiries, but it should also improve the customer experience.
Useful measures may include successful self-service resolution, repeat contact, escalation rates, customer satisfaction, time to resolution and the proportion of conversations requiring additional intervention.
Operational metrics such as contact volume and handling time remain important, but they should be interpreted alongside customer outcomes.
A business may reduce live-agent demand while unintentionally increasing frustration among customers whose enquiries are poorly suited to automation.
Similarly, a higher escalation rate is not necessarily a failure if it reflects more effective identification of customers who genuinely need specialist assistance.
The objective is to understand where automation helps, where it creates friction and what needs to change.
Questions to Ask Self-Service and Chat Providers
Before selecting technology or operational partners, contact centre teams should consider:
- Which customer enquiries are best suited to self-service?
- Does the solution support knowledge bases, guided journeys, chatbots and live chat?
- How does the system access and validate customer-service information?
- Can AI-generated answers be restricted to approved knowledge sources?
- How are unresolved enquiries identified?
- Can customers easily request human assistance?
- What information transfers when a conversation is escalated?
- How does the platform integrate with CRM and contact centre systems?
- Can it complete approved transactions and workflows?
- How are authentication and access permissions managed?
- What accessibility capabilities are available?
- How are customer conversations and personal data protected?
- Can managers review failed or abandoned interactions?
- How is successful resolution distinguished from containment?
- What implementation, training and ongoing optimisation support is provided?
Frequently Asked Questions
What is web self-service?
Web self-service allows customers to find information or complete tasks through digital channels without requiring direct assistance from a human adviser. It can include knowledge bases, guided processes, customer portals and AI-powered chatbots.
What is the difference between a chatbot and live chat?
A chatbot provides automated responses or actions using predefined rules, AI or a combination of technologies. Live chat connects the customer with a human adviser through a digital messaging interface.
What is conversational AI in customer service?
Conversational AI uses language-processing technologies to interpret customer requests and generate responses or initiate actions. Capabilities vary considerably between platforms and depend on their knowledge sources, integrations and controls.
How should self-service success be measured?
Organisations should consider whether customers successfully resolve their enquiries, alongside containment, repeat contact, escalation, customer satisfaction and operational efficiency. Containment alone does not prove resolution.
Should every customer enquiry be automated?
No. Automation is most useful where interactions are sufficiently predictable and can be resolved accurately and safely. Complex, sensitive or high-risk enquiries may require human judgement and intervention.
Product & Services Guide
Capacity
AI-powered customer-support platform providing knowledge management, virtual agents, helpdesk automation and agent-assist technology. Its solutions help organisations answer routine enquiries and support more efficient customer-service workflows.
Website: https://capacity.com/
Ciptex
Customer engagement technology specialist delivering cloud contact centre and communications solutions. Its capabilities include digital customer engagement, live chat, automation and integrations, with experience supporting organisations such as Shelter.
Website: https://ciptex.com/
ellio
Customer experience and business solutions partner combining operational delivery, advisory expertise and technology. It supports organisations managing complex customer interactions, including businesses operating in regulated sectors.
Website: https://www.elliocx.com/
Gnatta
Digital customer engagement and contact centre technology provider supporting organisations managing customer conversations and workflows across digital channels.
Website: https://gnatta.com/
HGS
Customer experience and business process management provider combining human customer-service operations with conversational AI, automation and agent-assist capabilities across multiple channels.
Website: https://hgs.com/
IP Integration Ltd
Contact centre technology and services specialist providing cloud communications, customer experience solutions and integration capabilities. Its portfolio includes technologies designed to connect customer interactions with wider contact centre workflows.
Website: https://ipintegration.com/
Kerv
Technology and customer experience specialist delivering cloud contact centre solutions, AI chatbots, systems integration and consultancy. Its capabilities include Genesys Cloud deployments and connected customer-service environments.
Website: https://kerv.com/
Magellan
Outsourced customer-service provider offering live chat support alongside wider contact centre and business process outsourcing services, including support for businesses seeking additional digital service capacity.
Website: https://www.magellan-solutions.com/
Wavenet
Managed technology and communications provider offering contact centre solutions, AI-enabled self-service, automation and integrated customer experience technology.
Website: https://www.wavenet.co.uk/
Zoom
Communications and contact centre technology provider offering Zoom Contact Center and Zoom Virtual Agent. Its capabilities include web chat, conversational AI and integration between automated support and human contact centre interactions.
Website: https://www.zoom.com/
Make Resolution the Measure of Success
An effective self-service strategy can be expressed through six stages: customer need → knowledge → automation → resolution → escalation → improvement
Understand why customers make contact.
Provide accessible, accurate information.
Automate interactions where doing so adds value.
Measure whether customers actually resolve their enquiries.
Make escalation straightforward where human help is required.
Use the resulting information to improve the experience.
The strongest self-service operation is not necessarily the one with the highest chatbot containment rate.
It is the one that resolves straightforward enquiries efficiently while making more complex interactions easier for customers and agents alike.
The Contact Centre & Customer Services Summit brings together senior contact centre decision-makers and relevant solution providers through pre-arranged one-to-one meetings, providing an opportunity to explore web self-service, AI chatbots, live chat, customer experience technology and operational support.
Related Reading
This is the first article in our October Web Self-Service & Chat series.
The follow-up will focus on improving performance after implementation: how contact centre teams can increase genuine self-service resolution, identify failed customer journeys, reduce chatbot frustration and make the transition to human assistance more effective.
Sources
- ICO – Guidance on AI and Data Protection: https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/artificial-intelligence/guidance-on-ai-and-data-protection/
- GOV.UK – Understanding Accessibility Requirements for Public Sector Bodies: https://www.gov.uk/guidance/accessibility-requirements-for-public-sector-websites-and-apps
- W3C – Web Content Accessibility Guidelines: https://www.w3.org/WAI/standards-guidelines/wcag/
- Capacity: https://capacity.com/
- Ciptex: https://ciptex.com/
- ellio: https://www.elliocx.com/
- Gnatta: https://gnatta.com/
- HGS: https://hgs.com/
- IP Integration: https://ipintegration.com/
- Kerv: https://kerv.com/
- Magellan: https://www.magellan-solutions.com/
- Wavenet: https://www.wavenet.co.uk/
- Zoom: https://www.zoom.com/
Image credit: https://unsplash.com/photos/blue-and-white-logo-guessing-game-EefRxCpIxnA
