A Hero image of a chatbot on an ipad

8 Reasons You Need a Chatbot For Your Website

The recent surge in customer expectations – which might be attributed to the fast rate at which we can receive and consumer things – married with externals forces such as the Covid-19 pandemic and consequential staying at home, has driven the demand for self-service options.

But do you need a chatbot for your website?

A report by Gartner estimates that by 2023, more than 60% of all customer service engagements will be delivered via digital and web self-serve channels, up from 23% in 2019.

This article explores the importance of including a chatbot in your customer service offering, its benefits and how it complements other key tools.

With chatbots estimated to have reached the early majority stage of the technology innovation curve, it is critical that in order to satisfy customers’ expectations and remain competitive in a time when customer service has never been more important, you consider a chatbot for your website.

An image that shows where chatbots currently reside on the technology adoption curve.

1. Significant Reduction of Contact

Chatbot technology that is built on AI harnesses the power of Natural Language Processing (NLP) so that large volumes of routine queries can be automatically dealt with, simultaneously and at scale.

An image that show a chatbot dealing with routine queries whilst it escalates non-routine queries to live chat.

NLP uses sophisticated algorithms to unpick the structure of a customer query, analysing each component including, keywords, grammar, search intent and popularity in order to understand the context of what is being asked. NLP then matches the customer query with the most relevant results, producing options for customers to choose from.

With the capability to reduce average contact volume by 25% , far fewer queries end up reaching the contact centre – because AI has handled them. The result of which is a reduction in not only the congestion experienced by contact centres but also the accumulation of costs involved with handling customer queries. Costs that surround handling queries directly, staffing and operational overheads are reduced therefore bringing overall contact centre costs down considerably.

Another key area that a chatbot’s self-service capability has a positive impact on is customer service and your CSAT and NPS ratings. With 67% of customers preferring to self-serve over speaking to a company representative, considering a chatbot for your website could provide huge CX benefits.

Further, when chatbots deal with a large proportion of routine contact, less reach your agents. The result is hugely beneficial for customers who have complicated issues that can only be solved by a human, there is a reduced wait time and they have agents’ full attention. This is because the reduction of routine queries gives agents greater bandwidth to which they can concentrate on complex, non-routine queries.

2. Chatbots Are There for Your Customers at Any Hour

Convenience is vital when it comes to customer service, particularly for self-serving customers. The channel should be straightforward, easy to navigate and available whenever your customers require help.

Chatbots offer real-time support 24/7/365. Customers needn’t worry when they have an issue during a public holiday or outside of traditional operating hours because a chatbot is entirely self-sufficient when it comes to offering support, relying on AI only to deliver results.

This proves particularly important for those who require help during emergency situations, for example, a water leak that takes place during the early hours of the morning. By reporting this to your website’s chatbot, it can offer instructions as to how this can be fixed.

3. Chatbots Boost Lead Generation and Sales

In addition to customer service, chatbots support both lead and new revenue generation activities. Acting as a lead generation tool, chatbots on your website can capture key contact information and qualify prospects before passing them onto the Sales team. By capturing critical data including personal, preferences and search intent, the lead can be escalated to an agent who can begin the nurturing process.

This also helps to identify many hot leads that would otherwise have been lost, anonymous website visitors.

To promote additional revenues, custom triggers are configured to intercept website visitors that have met certain conditions, for example, a pattern that suggests a high purchase probability. By being there to proactively offer assistance the chances of purchase increase.

4. Escalation to Agents Is Seamless

Channels such as chatbots that facilitate self-service are hugely beneficial however, it is important to consider that not all queries can be dealt with using AI. Some that are non-routine require human intelligence and empathy. This is where a chatbot’s capability to seamlessly escalate to an agent-assisted channel such as live chat is valuable.

When such escalation points are not available to customers it means they must search through pages and search engines results for a way to contact an agent, potentially wait and then repeat their query. This proves frustrating for customers, especially when you consider their calibre of the query is urgent. It affects not only First Contact Resolution (FCR) rates negatively but also lowers CSAT scores.

5. Chatbots Collect Knowledge and Integrate with Other Tools

Not only can chatbots integrate with other key customer service tools such as live chat, sharing data seamlessly within the same window, but they can also help inform and build your central repository of knowledge. Chatbot software that includes comprehensive analytics reveals intel by identifying any patterns or trends in customer queries and results. When analysed, this data becomes knowledge that is eventually utilised across the company.

Furthermore, when selecting a chatbot for your website it is important to consider its integration capabilities. Consider chatbots that use open RESTful APIs to ensure that integrations between other crucial business tools like CRMs and email management apps are available for 2-way data sharing.

6. As Digital Concierges, Chatbots Enhance CX

Chatbots can transform the experience that your customers receive. Acting as their digital concierge, the chatbot guides your customer through their online journey from start to finish, until they have reached their destination.

This is achieved by utilising decision tree technology that when configured, offers your customers a series of questions, that depending on the answers selected, determines the next step in their journey. This helps customers reach their destination far quicker and helps by automating the problem-solving process for them.

Chatbots enhance CX further by utilising Machine Learning (ML) principles. By recognising linguistic and behavioural patterns that can be stored and used for future interactions, things become more familiar and enjoyable for the customer.

7. Chatbots Channel the Essence of Your Brand

A chatbot for your website is not only a means for customer support, CX and revenue generation, it is also a way to channel your brand personality to the world. It’s a branding opportunity that is seen by every site visitor.

Configure your chatbot so that it reflects your brand personality including the responses it gives, the grammar it utilises and the way it looks. Some chatbot solutions include additional search layers that are configured to ensure that a conversational response is always produced.

8. Chatbots Are A Hit with Your Agents

Although not directly linked to the contact centre, the indirect impact that chatbots have on agents is hugely beneficial. Because chatbots take care of a large portion of routine queries, agents are able to focus on more complex customer issues, bringing them job enrichment, purpose and empowerment.

Providing agents with the opportunity to develop their skills and grow builds employee satisfaction (ESAT) and can be contributed to greater outcome and quality of work. For improved efficiency and agent productivity, it is therefore worth considering a chatbot for your website.


If you enjoyed this article and would like to know more about chatbots, read our guide here or for help regarding your organisational requirements, please

An image of a piggy bank

The 8 Biggest Chatbot Mistakes That Are Costing You Money

Whilst AI can be dated back to the 1950s when Turing famously invented the Turing Test, the first ever chatbot, ELIZA was introduced in 1966, followed by a series of ‘chatterbots’ in the 1990s. This wave of chatbot technology was not created to support customer service, rather they were tested by audiences to develop bot intelligence.

Today, chatbots are considered a must-have tool belonging to the wider customer service toolkit, facilitating not only customer support but lead generation, sales and growth. Therefore, it is not just customer service teams that value chatbots but also Sales and Marketing.

However, when it comes to selecting chatbot software and the subsequent implementation process, a surprising number of companies get it wrong. The results of which can be detrimental and include unnecessarily high overheads, poor CSAT and damaged brand reputation.

So, how can you ensure that the chatbot solution you select and your roll-out method works effectively for your company, achieving goals that surround:

  • Support cost reduction
  • CSAT and NPS improvements
  • Enhanced CX
  • Contact centre efficiency
  • Lead and revenue generation

This article addresses the common mistakes that all too many companies make when it comes to chatbots, offering advice as to how they can be avoided:

  1. Neglecting Objectives and Strategy
  2. Not Working with A Vendor Experienced in Your Field
  3. Not Giving Your Chatbot a Personality
  4. Selecting A Chatbot That Cannot Provide the Right Answers
  5. Choosing A Solution That Cannot Escalate to A Human
  6. Making A Decision Based Solely on Pricing
  7. Choosing A Chatbot That Has A Lengthy Implementation Process
  8. Settling for A Chatbot That Cannot Be Optimised

1. Neglecting Objectives and Strategy

Common mistake:When companies do not discuss and define clear goals with measurable objectives regarding their upcoming chatbot initiative, a multitude of things go wrong. The result of which means that those involved have no clear direction as to what they’re working towards, with no benchmark to know if they are on track or if something has gone wrong. Without objectives, there is nothing to measure the success of the outcomes against.

Best practice:Meet with everyone who will be involved with the deployment of your chatbot solution to ensure that everyone is on the same page with goals, expectations and outcomes. Not only does this allow for smoother execution but it helps to identify issues in their tracks. Having input from those involved provides a varied perspective, providing value to key decisions regarding chatbot selection.

2. Not Working with A Vendor Experienced in Your Field

Common mistake:Many businesses choose chatbot software vendors that are inexperienced in the field in which they operate. Chatbots that have not been tested in a certain field lack intent sets and therefore do not know their audience. The result of which includes money and time spent with vendors who cannot produce the chatbot required to fit your customers’ needs.

Best practice:Spend time and care when considering chatbot vendors. Compile a list of must-have criteria for your chatbot; its functions, capabilities and vendor experience. Ensure that the vendors that make it to your shortlist have sufficient experience in your field and can demonstrate tangible examples and results. Their experience will prove hugely advantageous during implementation.

3. Not Giving Your Chatbot a Personality

Common mistake:Neglecting your chatbot’s personality results in a forgettable CX and denotes a boring brand reflection. Chatbots that are missing a name, icon, personality and tone of voice will appear outdated. Whether your chatbot solution is incapable of such configuration or it has been neglected, the result will not be positive for CX.

Best practice: Ensure that the chatbot solution you choose offers flexible configuration; enabling the bot’s responses, language and grammar to match your brand’s personality and tone of voice. Certain chatbots include additional configurable search layers that ensure a conversational, on-brand response is always served.

4. Selecting A Chatbot That Cannot Provide the Right Answers

Common mistake:When companies opt for basic chatbot solutions that are not powered by AI or utilise Natural Language Processing (NLP), customer satisfaction suffers significantly. This is because the bot relies on customer queries matching its records exactly in order to produce responses. Because it is highly unlikely that every customer enters the chatbot’s exact records, a correct result is rarely produced. Instead, the chatbot replies with “I’m sorry I don’t understand your question. Please try again.”

Best practice:Always choose chatbots that are built on AI and harness the power of NLP. This ensures that regardless of how a customer phrases their query, a relevant answer will be produced. The way this works is through NLP’s sophisticated layers of search. By unpicking the query structure, it analyses each component including keywords, grammar, intent and popularity to understand context and therefore return relevant results.

To find out more about the difference between simple and AI-powered chatbots, click here.

An images to represent the 4 layers of Natural Language Processing: search keywords, intent, grammar and popularity.

5. Choosing A Solution That Cannot Escalate to A Human

Common mistake:Whilst chatbots are effective in dealing with routine questions, there will always be customers who require help with issues that are more complex and warrant human rather than artificial intelligence. When a chatbot cannot firstly identify when escalation is required and secondly, cannot provide a smooth escalation to an agent, then CX and CSAT plummet. When customers have serious issues they need resolving, having to repeat the process and their query over and over again is not ideal.

Best practice:Ensure that the chatbot solution you select offers seamless escalation options. Make sure that the chatbot can detect when it is not capable of resolving the query and that escalation to other agent-assisted channels such as live chat is stress-free, taking place in the same window. It is also key to the customer’s experience that transcripts are carried over to avoid unnecessary repetition, reducing Average Handling Times (AHT).

An example of a live chat widget

6. Making A Decision Based Solely on Pricing

Common mistake:Too many companies focus solely on finding the cheapest chatbot solution and unfortunately have to pay the price in the long run. Such solutions might seem tempting but generally end up costing you more over time. This is because lower-priced chatbots don’t utilise NLP which leads to the laborious process of manually programming lots of questions and answer pairs. Further, poor escalation provided by such solutions result in higher contact volume reaching contact centres and therefore unnecessary high support costs.

Best practice:Ensure that during your chatbot software selection process, you stick to a comprehensive list of essential requirements. Include attributes such as NLP utilisation and escalation capabilities that facilitate the escalation from chatbot to cost-effective agent-assisted channels such as live chat. This not only improves CX but minimises telephone and email queries which prove significantly more expensive to handle.

7. Choosing A Chatbot That Has A Lengthy Implementation Process

Common mistake:Commonly, the software implementation process is not discussed with the vendor. This results in companies waiting significantly longer than anticipated to have their chatbot up and running. When expectations are not discussed, the implementation process can become lengthy and this affects your customers and operational costs negatively.

Best practice:Always make sure to discuss and agree on the chatbot implementation timeline with your vendor. This avoids any confusion, manages expectations and ensures you will be up and running on time. It is also crucial to choose a chatbot that utilises low-code implementation, this makes setup simple, using one line of code and minimising the need for cross-departmental involvement.

8. Settling for A Chatbot That Cannot Be Optimised

Common mistake: Companies that opt for a chatbot solution that does not include a built-in analytics function removes the opportunity for optimisation. Without analytics that demonstrate how well your chatbot is performing in terms of resolved queries, escalation and trigger management there is no way to identify issues or plan for improvements. In other words, whilst competitors are constantly optimising their chatbots, yours remains stagnant.

Best practice:Ask your prospective chatbot vendors for a demo. Ensure that one of the areas that they show you is the software’s analytics function. Choose a chatbot that offers a comprehensive range of visual analytics that can be downloaded or viewed in a dashboard aesthetic. Such metrics will help you to identify any gaps or errors that emerge and focus on areas for optimisation and improvement so that your chatbot is working to its full potential.


If you enjoyed this article and would like to know more about chatbots, read our guide here or for help regarding your organisational requirements, please

Image of man using a chatbot on an ipad

Conversational Chatbots: The Fundamentals For CX

Chatbots: Taking CX By Storm

Just decades ago, chatbots were considered futuristic or gadget-like, they were innovations with a huge untapped potential for CX. The chatbots we are familiar with today, however, are functional customer service tools that have taken CX by storm, particularly in recent years.

For many businesses, chatbot are now deemed essential – if they aren’t already part of the existing technology stack, they are quickly making their way onto CX roadmaps across industries. According to one study, 77% of executives have already implemented and 60% plan to implement chatbots for after-sales and customer service.

For customers, chatbots provide familiarity, convenience and instant access to relevant information on your company, products or services. This not only enhances CX but drives demand as the global chatbot market is expected to grow from $2.6 billion in 2019 to $9.4 billion by 2024 at a CAGR of 29.7%. For companies who wish to remain competitive but are yet to implement chatbots into their current offering, they are worth considering.

Simple Vs Conversational Chatbots

There are major differences between simple and conversational chatbots that can affect your customers considerably. Whilst simple chatbots often seem the more cost-effective option, when it comes to fulfilling your long-term CX strategy, this is where they fall short.

Types of Chatbot

There are 3 common chatbot types used in customer service, these include:

  • Menu-based chatbots
  • Keyword recognition chatbots
  • Conversational chatbots

Menu-based
chatbots

Keyword recognition chatbots

Conversational
chatbots

An image to demonstrate the differences between basic and advanced AI chatbot software.

Menu-Based Chatbots

Using menu buttons to help customers navigate to the answer required, this chatbot type has basic functionality. Menu-based chatbots are built on rule-based automation as opposed to AI, which means that they can only respond to queries that match their pre-loaded responses exactly. The limitation here is this chatbot will not recognise even the smallest query variation, this results in a dead-end response without the capability to further attempt to understand what a customer is asking.

Whilst this simple chatbot type might be adequate for very small or start-up businesses with extremely basic needs, for example serving a fixed set of FAQs, it is not sustainable for larger companies with more sophisticated CX goals.

Keyword Recognition Chatbots

This type of chatbot utilises basic AI to break down the query at hand, analysing each keyword to deliver the most relevant result. By focusing on different word classes, keyword recognition chatbots can determine the most suitable response. For instance, by isolating keywords “renew” and “subscription” or “setup” and “account”, the chatbot can assume the customer’s requirement and send a response based on this.

Although keyword-recognition chatbots harness AI to some extent, they are not effective at recognising and conversing with multiple query variations.

Conversational Chatbots

When it comes to delivering CX, conversational chatbots are by far the most effective type of chatbot. These advanced tools utilise AI, harnessing Natural Language Processing (NLP) to understand the context and intent of the question that is asked. This means that multiple variations of the same query can be asked and an identical answer is delivered seamlessly. Even if a question is not immediately obvious, conversational chatbots use decision tree technology to ask a series of questions until a resolution is found.

Conversational chatbots will never serve the response: “I’m sorry I don’t understand the question. Please try again.”

Unlike its basic alternatives, this chatbot type can be configured to naturally converse with customers, adding character to the experience whilst conveying brand personality. Through machine learning principles the CX is further enhanced as customer journeys are personalised; conversational chatbots can learn, store and use customer information for future sessions. By remembering certain details and preferences CX is of optimal efficacy.

Basic, rule-based chatbot

An image to demonstrate the differences between basic and advanced AI chatbot software.

AI-powered chatbot

An image to demonstrate the differences between basic and advanced AI chatbot software.

Why Conversational Chatbots?

A report by Gartner reveals that 91% of organisations plan to deploy AI by 2022. Another report suggests that by 2025, 80% of large enterprises will need to have a “conversational-technology-focused-centre” implemented.

AI and the tools in which it powers are rightly viewed as game-changing technologies.

While AI has been around since the 1950s when Turing developed the Turing Test, followed by the debut ‘chatterbot’, ELIZA, that was developed in the 1960s, why is it that the conversational chatbots of today are experiencing such rapid adoption?

The answer is completely customer-centric; conversational chatbot adoption is customer-driven. As more and more businesses compete for customers, more and more new and innovative technologies are used to drive a high level of service. As they become accustomed to this, customers expect more from businesses in general. They want less effort and a better experience. They want the ease and convenience of using an online contact channel and the depth and personality associated with agent-assisted channels.

Chatbots do not only provide a familiar instant messaging interface for customers, contributing to an enjoyable experience, but they assist in guiding customers through their journey, from start to finish. When configured correctly, chatbots are your customer’s first port of call when looking for advice, they can be guided to their destination in a number of ways:

  • Most commonly, conversational chatbots solve customers’ routine questions instantly using AI-powered automation
  • For instances when customers don’t yet know the resolution they require, but know there is a problem, conversational chatbot utilise decision tree technology to take care of the problem-solving process until the correct resolution is found
  • Chatbots detect when a query is non-routine and therefore escalates the customer to an agent-assisted channel such as live chat where the adequate help can be provided – all of which takes place in the same window

Conversational chatbots are not only a hit with customers but with customer service and contact centre teams alike. Their capability to automatically handle significant contact volumes allows agents to focus on the queries that are complicated by nature, boosting CSAT and agent satisfaction. As a Result, Average Handling Times (AHT) are reduced by 25% and First Contact Resolution (FCR) is increased by 80% (Synthetix research).

NLP, NLU, NLG & Machine Learning

How do these products of AI impact conversational chatbots and CX?

Natural Language Processing

Natural Language Processing (NLP) is a branch of AI that deals with linguistics, its main purpose is to help machines understand the human language. NLP comprises Natural Language Understanding (NLU) and Natural Language Generation (NLG) to naturally interact with customers, simulating a real conversation.

Natural Language Understanding

Natural Language Understanding (NLU) uses algorithms to isolate and analyse the contents of a customer query. By identifying word classes and detecting sentiment, topics, entities and intent, NLU is essentially capable of comprehending context and what a customer is asking.

Natural Language Generation

Natural Language Generation (NLG) is the process of taking the structured data that has been produced as a result of NLU and transforming it into consumable, natural language. Algorithms that understand the construct of a naturally phrased sentence build responses based on the understanding and processing of the interaction.

Machine Learning

Machine Learning, a subset of AI surrounds the idea that computers can automatically learn and improve based on experience opposed to human intervention. Conversational chatbots adopt Machine Learning principles to personalise and enhance CX. By identifying trends in customer information, storing it and them remembering it for future interactions, chatbots create a positive and efficient experience for customers.

Implementing Conversational Chatbots

How do these products of AI impact conversational chatbots and CX?

At first glance, the implementation of conversational chatbots might seem daunting, but with the correct tools, processes and support, it’s straightforward.

Following the vendor selection process and agreed SRS, your first step depends on whether you have an existing knowledge base or not. You will either:

A. Integrate seamlessly with the chatbot so that knowledge can be shared back and forth between the knowledge database and chatbot interface

B. Collect all existing and extract any new company knowledge in order to populate your centralised knowledge base – you can find out more about the knowledge collection process here

Conversation Chatbots - Image of chatbot being sourced by knowledge

Next, configure any secondary search layers that are responsible for understanding grammar and synonyms. For example, Synthetix’s system, “Jabberwocky” unravels customer query sentence structure to understand the meaning of any synonyms or quirks that your knowledge base is not familiar with. This ensures a conversational response is always delivered and increases accuracy. You can also configure this system to match your brand’s tone of voice so that personality is effectively conveyed during conversations.

Following successful implementation, it is good practice to closely monitor analytics for usage and trigger management data that can determine how effectively the conversational chatbot is working. Settings can be adapted and crucial decisions can be made based on such analytics for future CX improvements.


If you enjoyed this article and would like to know more about chatbots, check out our guide here. If you’d like any advice on conversational chatbots or help with implementation, please

Contact Centre Chatbots: Why Are They Critical to CSAT?

Why Chatbots?

Whilst they have been around since the 1960s, when the first chatbot, ELIZA was born, it wasn’t until recent years that the adoption of chatbots within business exploded. With the utilisation of AI and other intelligent functions, chatbots are no longer considered ‘futuristic’, rather a key component of the customer service ecosystem.

The global chatbot market is expected to exceed more than $10 billion by 2026 and will grow at a CAGR of more than 23% in the given forecast period.

It is also predicted that by 2022, chatbots will have saved businesses $8 billion per year, which is one of the key drivers for adoption and a fundamental reason why contact centre chatbots are fast becoming essential business tools.

So, how do they work?

Chatbots are a form of self-service that offer customer support through a familiar direct messaging interface. Built using AI and utilising Natural Language Processing, chatbots can facilitate a two-way conversation whereby the customer enters their query and the chatbot asks qualifying questions so that the most relevant answer can be delivered.

NLP utilises Natural Language Understanding (NLU) to dissect the query into keywords, search intent, grammar and popularity. Essentially understanding query context, an adequate and relevant answer can be generated through Natural Language Generation (NLG) using a conversational manner that matches your brand personality.

A image to demonstrate the 4 layers of search intent that is used by Natural Language Processing at Synthetix.

These mechanics help chatbots to understand multiple variations of the same question, meaning that they can successfully answer a query regardless of the language, grammar or idiosyncrasies that are entered. Chatbots that do not utilise NLP rely on customer queries matching their articles exactly in order to trigger a resolution, otherwise unhelpful responses such as, “I’m sorry, I don’t understand the question.” are delivered.

Basic, rule-based chatbot

An image to demonstrate the differences between basic and advanced AI chatbot software.

 

AI-powered chatbot 

An image to demonstrate the differences between basic and advanced AI chatbot software.

Adopted by a multitude of businesses across many industries, chatbots can:

  • Reduce Average Handling Times (AHT)
  • Increase First Contact Resolution (FCR)
  • Cut operational costs
  • Boost CSAT
  • Enhance CX
  • Promote agent efficiency
  • Improve employee morale

Further, the role of the contact centre chatbot within the wider customer service ecosystem is an important one. Acting as an online concierge, it handles routine questions and tasks, whilst providing escalation to agent-assisted channels like live chat when human intervention is required.

How Do Chatbots Benefit the Contact Centre?

No longer a perceived threat to contact centres, chatbots are now considered a vital tool for agents. Working in amalgamation with chatbots helps to combat the growing volume of routine contact queries associated with the explosion of digital transformation.

Contact centre chatbots have become increasingly popular in recent years, particularly in 2020 as a result of the pandemic. With a surge in support calls and emails and some organisations experiencing up to a 133% spike in contact , chatbots provide support to agents working remotely and in the contact centre.

Chatbots Help Your Contact Centre Boost CSAT

Through continuous AI improvements and NLP development, chatbots have evolved to better understand the context and intent of the customer queries they are asked. With advanced AI, chatbots can not only handle routine queries but many routine tasks, such as:

  • Taking a meter reading
  • Submitting a claim
  • Resetting a password
  • Booking and confirming a reservation
  • Checking a delivery’s location

Your chatbot can even be configured to match a certain tone of voice, providing the personal experience that so many customers require.

Whilst agents can only handle 1 call or email at a time, chatbots can automatically deal with large scale routine queries and tasks simultaneously. This, therefore, significantly frees up any channel congestion caused by large volumes of routine questions – the result of which means more customer queries are effectively dealt with and consequently, CSAT scores improve.

It also frees up agent resources, creating time and space for agents who can now provide their full attention and expertise to customers who require urgent attention due to their complex issues. If a customer is in a vulnerable position and needs urgent assistance, the last thing they need is to be waiting in a long queue to speak to an agent. This only reflects negatively on CX and CSAT but can be avoided by utilising contact centre chatbots.

An icon of an light bulb

Many customers prefer to self-serve if they can – around 73% in fact. By including a chatbot in your customer service offering you cater to your customers’ needs, improving CSAT.

Further, chatbots provide customers with 24/7/365 support, they can resolve queries and instruct customers outside of contact centre operating hours or during public holidays. AI chatbots can even assist customers through emergencies in real-time, successfully delivering relevant emergency contact details and instructions required.

Chatbots Contribute to Agent Satisfaction

Contact centre chatbots, contrary to popular belief, don’t hinder agents, rather they have the capabilities to enhance their quality of work and job satisfaction.

When the vast majority of queries that come through the contact centre are the same mundane routine questions and monotonous tasks, this can take a toll on your agents – causing drive and morale to decline.

However, these routine questions can be dealt with at scale through AI, using chatbots, creating a greater bandwidth for your agents to which they can deal with more complicated, non-routine issues. When agents are able to dedicate their attention to helping customers with serious or complicated enquires it adds variety and purpose to their roles, which in turn contributes to job enrichment and motivation.

It’s no revelation that when employees are happy, this is reflected positively through their quality of work. So, when chatbots are introduced into the ecosystem, agents not only develop skills quicker but they are supported and happier at work, which has an impact on the service that customers receive, enhancing CSAT.

Chatbots Make Your Contact Centre More Efficient

Many company’s customer service and contact centre operations lack efficiency, particularly when it comes to the mechanics of the contact channels. Too often expensive agent-assisted channels are unnecessarily exhausted, or low-cost AI-powered channels are not utilised or difficult to find.

This is where chatbot solutions can assist the contact centre’s efficiency. Ultimately, when configured correctly, chatbots can sort queries into the correct channels, avoiding unnecessary costs. For example, if a chatbot is configured to be a customer’s first port of call for customer support it can either:

A. Deal with the customer’s routine query successfully

B. Or, detect that the customer’s query is non-routine and requires human assistance, automatically escalating the customer to an agent-assisted channel such as live chat

Through automation, chatbots can deal with routine questions significantly quicker than humans. There are many time variables when it comes to manually opening, replying, researching and resolving just one customer query compared to AI handling everything. This cuts your operational costs considerably, accounting for up to 30% support cost reductions.

A chatbot not only escalates a query when it is non-routine, but it can also be configured to escalate to an agent-assisted channel based on the keywords the customer enters. Trigger management settings enable this escalation to take place based on certain keywords, for example, those that indicate urgency, such as “renew” or “cancellation”. By putting these customers in touch with a human that can efficiently deal with the matter, customer churn is reduced and revenue is increased.

Pair this with chatbots’ intelligent smart forms integration which collects targeted information from the customer throughout their journey and significant time is saved, slashing overall AHT.

Further, contact centres can become more efficient by utilising the data that chatbot technology produces. With detailed analytics revealing what customers are asking and article effectiveness, you can be in the know about current customer trends or issues. As a result, you can proactively better equip agents, building such intelligence into agent scripting and other learning tools.


If you enjoyed this article and would like to know more about chatbots, you can read our guide, here. or, If you would like any advice or assistance with contact centre chatbots

An image showing xan on a phone

What are Chatbots?

An Introduction to Chatbots

A chatbot is a software application primarily used by businesses to facilitate customer service. Unlike agent-assisted contact channels such as live chat or telephone, chatbots rely on AI to resolve customer queries. The tool that enables human and machine interactions has become increasingly popular over the last decade, often acting as customers’ first online port of call, but how did the chatbot become so sought after?

The first chatbot, ELIZA was created in the 1960s before the term ‘chatbot’, or its predecessor, ‘chatterbot’ was invented. ELIZA was an advance in computer science but was not made with the intention to assist customer service.

Following ELIZA, there were several failed attempts at producing an effective chatbot including the likes of Cleverbot and Tay AI. What these bots had in common was unregulated training data; the way in which they ‘learnt’ was from the information fed to them by their users – the public – opposed to being configured by professionals using benchmarking and objectives. The result of which led to Tay AI’s termination following a spurt of offensive views it shared on Twitter.

Today we are familiar with a different type of bot – the customer service chatbot. Unlike other chatbots whose purpose was simply novelty – to be played with and to provide entertainment – the chatbots we are used to today are there to provide help, automating customer queries, providing good CX and subsequently helping customer service teams.

How Do Customer Service Chatbots Work

Customer service chatbots are built on AI and through Natural Language Processing (NLP) and Machine Learning (ML) principles effectively ‘understand’ customer queries to provide adequate responses. Chatbots integrate with companies’ centralised knowledge so that depending on what query is entered, they can utilise smart algorithms to identify and serve the most relevant knowledge articles to customers.

NLP is the combination of Natural Language Understanding (NLU) and Natural Language Generation (NLG) and is constantly learning from each customer interaction to be as optimal as possible over time. NLP is responsible for unpicking the customer query, dissecting the sentence into intent-based keywords, subjects, grammatical quirks and keywords popularity to analyse, interpret and present an appropriate result.

An image to demonstrate the differences between basic and advanced AI chatbot software.

It’s important to note that not all chatbots offered today are powered by AI, some are built on rule-based automations. Such chatbots can only respond to queries that exactly match their pre-loaded responses, leaving no room for query variations or synonyms. The result of which is “I’m sorry, I don’t understand the question. Please try again” served to customers more often than a genuine result.

This is why AI-powered customer service chatbots are so valuable to business. They are programmed to understand that there are multiple ways that a question can be asked, an array of synonyms and even idiosyncrasies. NLP helps to eradicate unhelpful responses.

Some chatbot specialists include additional layers of search for chatbots to consult if query intent is unclear. Synthetix’s system, “Jabberwocky”, for instance, unpicks query sentence structure, analysing grammar and a range of word classes. Its utilisation of proprietary NLP enables brand personality to be portrayed, increases answer accuracy and ensures that the customer always receives a conversational response.

An image to show how Synthetix’s additional search layer, jabberwocky can compliment knowledge base articles for chatbots.

Why Use Chatbots In Customer Service?

According to IBM , chatbots in customer service can solve 80% of all routine queries – including questions such as:

  • What is your returns policy?
  • Where is my delivery?
  • Can I apply for a refund?

Such routine questions make up a large percentage of overall contact volume, so when they are automated through AI and solved at scale, operational costs are significantly reduced. Without chatbot solutions in place, customer service teams would otherwise deal with routine queries, creating a backlog of tickets and an accumulation of overheads and staffing costs.

30%

Did you know?
Chatbots can save businesses up to 30% in customer service costs.

It isn’t just the bottom line that benefits from chatbots’ capabilities to automate routine queries, agents are given a greater bandwidth in which they can deal with customer issues that are complex or sensitive by nature, rather than the same queries time and time again.
A study
 revealed that 64% of agents with AI chatbots are able to spend most of their time solving complex problems, compared to 50% of agents without AI chatbots. This in turn contributes to positive employee morale – their perceived quality of work is richer when the repetitive, mundane element is removed, promoting job satisfaction and therefore productivity.

Customer service chatbots have a key role to play in customer satisfaction. Not only does NLP ensure that customers are served the very best results, encouraging improvements in First Contact Resolution (FCR) rates, but the channel itself is easy to use and convenient. Offering 24/7, real-time results, it’s a channel that customers can depend on, if for example, a chatbot provides customers critical information in an out of hours emergency, this of course is reflected in CSAT scores.

The adoption of chatbots also helps to optimise CX through Machine Learning (ML) principles and decision tree technology. ML principles enable chatbots capabilities to learn, store and utilise trends in customer behaviour, for example popular queries, colloquialisms or preferences are collected and then used for future interactions making the overall CX more enjoyable.

Effective customer service chatbots provide a smooth and enhanced customer journey through decision tree technology. To reach a successful result, an agent will usually ask customers a series of questions, but when there is no agent involved, how is this facilitated? Decision trees are configured so that chatbots can intervene at the problem-solving stage and ask such questions. This helps customers reach their destination result quickly and effectively, seamlessly transferring them to agent-assisted channels like live chat if necessary.

Types of Customer Service Chatbots

There are 3 types of chatbots that are commonly used in customer service, these are:

  • Menu-based chatbots
  • Keyword recognition-based chatbots
  • Contextual chatbots

Menu-based chatbots use clickable menu buttons to help customers reach the result they require. Similar to the rule-based chatbot discussed previously in this article, this type of chatbot relies on one specific set of data and therefore cannot offer results based on anything outside of this. Although menu-based chatbots might be suitable for small businesses with super basic requirements, for instance providing answers to FAQs, they are simply not capable of answering queries more advance than this. As a result, the precision and efficiency of answers served suffer.

Keyword recognition-based chatbots utilise AI, isolating and analysing each keyword to determine the best results. With each query that is entered, this type of chatbot will concentrate on the keywords that warrant action, for example “order”, “account”, “payment” and verb word classes such as “setup”, “cancel”, “refund” to piece together the most suitable response for the question at hand. Keyword recognition-based chatbots, however, are not effective at recognising multiple variations of questions.

Contextual chatbots utilise both AI and ML principles, proving particularly valuable to customer service amongst a multitude of companies. This chatbot type learns and stores useful information to utilise in future conversations. By remembering certain customer preferences and behaviours, conversations with the same customer during different sessions are of optimal efficiency and CX. A large portion of time can be saved if the bot simply asks the customer if their preferences are the same this time around, opposed to carrying out the same question process time and time again.

In customer service, chatbots are not only an expectation but they can assist teams quite significantly whilst complimenting companies’ overall customer service toolkits. As chatbots are more widely accepted and businesses understand the importance of AI-powered tools, we can only expect bigger, more impressive things for the bots of customer service.


If you enjoyed this article and would like to find out more about chatbots,
you can do so here
 . Or, if you’d like any chatbot advice or help with your organisational needs, please

AI Chatbots in Customer Service: A History

An Introduction to AI Chatbots

In recent years chatbots have become an integral part of companies’ customer service offering, providing benefits such as:

  • Operational efficiency – the capacity to resolve mass queries at scale
  • Improved CSAT scores – utilising Natural Language Processing (NLP) for the best customer experience
  • Instant, 24/7 support – providing customers with around the clock assurance
  • Seamless customer journeys – guiding customers to the intended result
  • Brand personification – chatbots can be customised to represent company character
  • Lead generation – automating the qualification process for efficiency
  • Enhanced CX – Machine Learning principles enable improved experiences
  • Improved employee morale – automating routine queries allows agents to focus on complexities
  • Smooth escalation – identifying when a customer should be transferred to a human

So how does it work?

AI-powered chatbots utilise intelligent algorithms, Machine Learning (ML) principles and Natural Language Processing (NLP) to ‘learn’ from user behaviours and patterns to become more effective over time. Usually integrated with a centralised knowledge base, an AI chatbot can interpret query variations to identify the adequate result and deliver it using a conversational manner. The more interactions the chatbot has, the ‘smarter’ it becomes.

It is important to note that there is a key difference between an AI chatbot and a chatbot, one relies on artificial intelligence and the other relies on rule-based automations:

  • Basic, rule-based chatbots: built on simple FAQ automations, this type of chatbot can only respond to queries that exactly match its pre-loaded responses. These will only trigger when specific keywords are entered by the user, any variations will not register. The most common answer these chatbots serve is “I don’t understand the question”.
  • AI-powered chatbots: using multi-layered algorithmic Natural Language Understanding (NLU), these chatbots understand customer intent and respond conversationally using Natural Language Generation (NLG). This approach will deliver the right response, regardless of how a query is phrased.

Basic, rule-based chatbot

An image to demonstrate the differences between basic and advanced AI chatbot software.

AI-powered chatbot

An image to demonstrate the differences between basic and advanced AI chatbot software.

The global chatbot market is expected to grow by $1.11 billion during 2020-2024 – and CAGR of 29%. But how did we get here?

Early AI Chatbots

Artificial intelligence and bots can be traced back to as early as the 1950s when Turing famously designed the Turing Test which would determine whether a machine could pass as a human based on the answers it gave.

But let’s focus on 1990s when the then, “chatterbot” was born. Unlike the chatbots of today, the initial wave of chatterbots weren’t created to facilitate customer service. Instead they were essentially a toy that was played with to test the bot’s intelligence. The way these chatbots learnt was from people talking to them – the more intel their community fed them, the greater their database of ‘knowledge’ became.

Cleverbot (1997) and other chatterbots constructed their replies by drawing on their database of previous conversations rather than consulting a regulated set of knowledge articles through NLP. This quickly became problematic when it was made available to the public.

Why?

  • Because people were intentionally ‘teaching’ the chatbots offensive ‘knowledge’, it would serve users just that.
  • Due to the nature of the learning process, there was no control over the content that was being delivered, this meant no way to prevent profanity or inappropriate ‘views’ being shared by the bots.
  • These chatbots were viewed simply as toys, they learnt nothing of true value and didn’t serve a purpose other than to entertain.
A screenshot of the disclaimer featured on Cleverbot’s homepage.

Following this, there were several developments to try and introduce Machine Learning (ML) chatbots, including Microsoft’s Tay AI (2016). This wasn’t specifically focused on customer service like the chatbots of today but was Microsoft’s attempt to find out if a chatbot could learn like a human.

Tay AI was launched on twitter as a novelty bot, using the tweets she was sent as her training data, Microsoft explained: “The more you chat with Tay the smarter she gets.” The fundamental issue here was the uncontrolled training data – Tay was learning from users who were tweeting their offensive, inflammatory views and once she has drawn similarities and identified patterns, she began to repeat such statements. Her sentence structures were sound, but their content was nonsensical and often controversial.

This is an example of how unregulated training data doesn’t work for customer service chatbots, whilst she managed to learn punctuation and how to form a sentence, she could not identify that what she was saying was offensive. Tay AI backfired for Microsoft and was removed from Twitter after just 16 hours.

Modern Day AI Chatbots

Today, AI chatbots don’t use unregulated or generalised training data, instead they utilise carefully curated data based on customer needs. Such data is usually stored in a centralised knowledge base which is configured to contain ML within set parameters so that companies can reap the benefits of using a chatbot, whilst avoiding fiascos such as Tay AI’s.

Modern day AI chatbots are most commonly used in customer service and use NLP to deliver results in a conversational way. Once a user enters their query, intelligent algorithms and NLP analyse the query, dissecting the sentence into search terms, grammar, and keyword popularity to serve relevant articles to the user.

A image to demonstrate the 4 layers of search intent that is used by Natural Language Processing at Synthetix.

If there is not an obvious match available then some AI-powered chatbots use additional search layers such as Synthetix’s “Jabberwocky”, which utilises NLP to understand sentence structure. It carefully unpicks the sentence into word classes and identifies conversational responses based on the proprietary Natural Language Generation (NLG) which can also be manually edited and configured by the Knowledge Manager.

The purpose of such additional search layers is to inject brand personality into the conversation whilst improving the accuracy of answers provided. It also ensures that customers will always receive a conversational response, not “I’m sorry, I don’t understand that question” – which proves frustrating for customers, thus poor CX.

The AI-driven chatbots of today are significantly more controlled than those of recent decades with greater efficiency of customer service but with the same, engaging human-like interactions.

AI Chatbots In Customer Service

AI chatbots are considered a crucial part of a company’s online customer service offering and can prove invaluable to both customers and teams across a multitude of industries.

AI Chatbots For Customers

When executed well, chatbots are considered a valuable tool for customers, providing them with quick answers 24/7 through a convenient platform. This is especially useful for those in emergency, out of hours situations who require real-time access to contact details for assistance, or those who simply don’t have time to engage with agent-assisted channels.

AI chatbots can significantly improve customer experience through customer journey technology and ML principles. Certain chatbot solutions utilise decision tree technology to effectively guide the customer to their intended destination. Whether that be a knowledge article, product page or social channel, decision trees, that are carefully configured by a Knowledge Manager, ask a set of questions that will ultimately determine the customer journey and end goal. Without such measures in place, the chatbot assumes that the customer has carried out all necessary problem solving themselves – which is rarely true. The same technology can identify when it is best for an agent to intervene and can seamlessly escalate a customer to an agent assisted channel such as live chat.

An image to show Synthetix Decisions interface

Moreover, chatbots can learn from every customer interaction, storing popular keywords, grammar and colloquialisms for future conversations – constantly improving CX.

AI Chatbots For Teams

Customer service and contact centre teams can benefit significantly from introducing chatbots into their offering. Because they are powered by AI and configured to utilse NLP, the accuracy, efficiency and quality of answers delivered is of a high standard, this improves over time as bots learn from their interactions, having a positive effect on CSAT scores.

Chatbots allow routine questions to be automated, this results in mass queries being resolved at scale promoting not only operational efficiency but reducing overheads. Accumulated operational and staffing costs are cut because a large portion of queries are dealt with digitally and via a bot. Additionally, because routine queries are automated, contact centre employees are given greater bandwidth to deal with more complex, sensitive and subjective issues. This encourages job satisfaction, empowers employees and impacts positively on staff attrition rates.

Chatbots can also directly attribute to revenue streams. Acting as a lead generation tool, some chatbots are configured to qualify leads before they are passed on to sales teams. Equipped with data including the questions they have asked, the preferences they have chosen and using data capture functions, their contact details, leads can be instantly escalated to an agent to nurture. This is particularly useful for teams who previously would have no intelligence on potential ‘hot leads’, but through NLP, can be identified and moved into the sales funnel.


If you enjoyed this article and would like to know more about AI chatbots, you can read more here or