Joe O’Brien, Global Corporate Communications Manager at Sabio Group, takes a close look at the conversational AI tool, ChatGPT.

Over the last few weeks, you would have heard lots of noise about ChatGPT, the new model for conversational AI that was launched by OpenAI – the AI research and deployment company – at the end of November.

What is particularly striking about ChatGPT is that it took just five days to reach one million signed-up users, and it’s estimated that figure may already be over two million. In comparison, Instagram took three months to reach that number, Spotify five months, and Twitter two years.

So, what’s so exciting about just another conversational AI tool? Why is it clearly capturing so much attention? And what’s going on in the AI market, when just last month some commentators were questioning chatbot sector momentum – particularly with Amazon stripping costs and people out of its Alexa team?

What’s so different about ChatGPT? 

OpenAI has designed ChatGPT to be as accessible as possible, with a focus on optimising its language models for dialogue. This should make it possible for the model to answer follow-up questions, challenge incorrect statements, reject inappropriate requests and even admit its mistakes.

The ChatGPT model sits in the ’emerging generative AI ‘category and has been designed to offer human-like conversations across a range of diverse topics according to its domain-specific model building and training. Not limited to standard text interactions, ChatGPT can write poetry, help debug code and even troubleshoot IT issues. The way that the model is trained helps determine how questions are answered. Other OpenAI examples of generative AI include the DALL.E neural network that uses the same underlying GPT-3 technology to generate images from text descriptions.

However, unlike existing chatbots that interact with people in a ‘seemingly intelligent’ conversational manner, the GPT-3 neural network ML model enables ChatGPT to deliver responses that Gartner suggests: ‘appear to have understood the question, the content and the context‘ of the user’s question.

Unfortunately, models are only ever as good as their training. So just because ChatGPT might generate answers that come across as impressively coherent, there’s no certainty that they will always be correct!

Testing ChatGPT’s CX expertise 

So to find out more, I thought I would ask ChatGPT some questions from a CX and customer service perspective. Below are some verbatim excerpts from our ChatGPT session that certainly demonstrate coherence:

• Question 1.: How can ChatGPT help contact centres?

ChatGPT answer: ChatGPT is a powerful language model that is trained to respond to natural language inputs. In a customer service setting, ChatGPT could be used to assist human agents to answer customer inquiries and provide helpful information. This could help contact centers to improve the efficiency and accuracy of their customer service operations. For example, ChatGPT could be used to quickly provide answers to frequently asked questions, freeing up human agents to focus on more complex or specialized inquiries. It could also be used to provide personalized responses to customer inquiries, improving the customer experience. However, it’s important to note that ChatGPT is a tool and should be used in conjunction with human oversight and judgment to ensure that the information provided is accurate and helpful. 

• Question 2.: Does ChatGPT work with Conversational AI?

 ChatGPT answer: Yes, ChatGPT is a type of conversational AI. It is a large language model that is trained to generate human-like text responses to input. This makes it well-suited for use in conversational AI applications, such as chatbots or virtual assistants. ChatGPT can be integrated with other conversational AI systems to provide more natural and accurate responses to user inquiries. This can help improve the overall user experience and make it easier for users to interact with the AI system. However, it’s important to note that ChatGPT is a tool and should be used in conjunction with other technologies and human oversight to ensure that the responses it generates are accurate and helpful. 

• Question 3.: How could contact center agents take advantage of ChatGPT?

ChatGPT answer: Contact center agents can take advantage of ChatGPT in a number of ways. For example, ChatGPT could be used to quickly provide answers to frequently asked questions, freeing up human agents to focus on more complex or specialized inquiries. This could help improve the efficiency of the contact center and reduce the time that agents spend on routine tasks. ChatGPT could also be used to provide personalized responses to customer inquiries, improving the customer experience. This could be especially useful for handling high volumes of customer interactions, as ChatGPT could help agents to respond to a larger number of inquiries without sacrificing the quality of the responses. Additionally, ChatGPT could be used to provide real-time assistance to human agents, helping them to generate more accurate and helpful responses to customer inquiries.

Interesting, but caution is key…

Initially these responses look great – addressing the question, coming across as knowledgeable and informative, and also introducing a note of caution about the bot’s limitations – as well as the requirement for human oversight and judgement. One analyst described ChatGPT answers as ‘looking like a student trying to answer a question without having attended the lecture‘. Indeed, this veneer of confidence could easily convince users that they’re accessing expert insight without realising that it’s often just well-crafted tech-speak.

Although it’s correct to be cautious, this doesn’t mean that a generative AI model such as ChatGPT will never be ready for enterprise CX deployments.

It’s more a recognition that this is an AI solution that’s barely out of the lab. The fact that it’s so easy to use and potentially so powerful would suggest that AI and automation will only have a more significant role to play in shaping the customer journeys of the future.

How that will evolve will depend on how effective generative AI tools become in terms of enabling more cost-effective and intelligent chatbots and web customer service solutions…

In the meantime, you can try ChatGPT for yourself by registering for OpenAI.

Alternatively, you can learn more about how the contact centre, AI & Automation and CRM landscape is starting to converge by downloading Sabio’s White paper here.

In addition, we’ll be taking a look at this subject in more detail at our flagship digital transformation event, Disrupt. To register for the event and to hear our session – ‘Open AI and Chat GPT; What it means for CX’ visit our website here.

About the Author

Joe O’Brien is Global Corporate Communications Manager at Sabio Group.

Recent news articles are telling us that the world of voice assistants is struggling, with the latest industry murmurings involving Amazon’s Alexa and it being described as a “colossal failure” rumoured to be losing $10 billion per-year. Phil Jordan, AI & Automation Specialist at Sabio Group, investigates.

It’s hard to imagine that this, now common, household item of a voice assistant is so hard to monetise, given how useful they are. I myself am the owner of several Google home devices, and I use these multiple items every day to perform various tasks.

At various times throughout the day, a variety of commands can be heard. “Hey Google, play the Beastie boys” for some entertainment or “Hey, Google, set a timer for 12 minutes” when I’m cooking.

For some practical use, I may throw in some commands such as “Hey Google what’s the weather like in London today?” or “How long will it take for me to get to Nottingham?”

Then there’s a little parenting…

“Hey Google, turn off the PlayStation” closely followed by “GET YOUR SHOES ON AND GET IN THE CAR!!!” (not at Google, I may add)

And then there’s even time for some homework “Hey Google, what on earth is a “conjunction”?”

The fact is though, once the hardware has been sold, in Amazon’s case regularly at a cost price, it is difficult for big tech companies to monetise any of these transactions. Google has tried to reinvent this view by creating the Nest product suite, which although comes at a higher price, also has arguably better sound quality and the ability to link multiple devices for stereo quality for a better music streaming experience, (since this is what people were commonly using the devices for). The only way they would look to monetise this functionality is by offering un-skippable ads for every transaction, but this would quickly make the device unusable, as it is a tool for performing a single specific function rather than an interface which I am using for entertainment or research purposes (or parenting!).

Added to this, the maintenance of such language models on a global scale, supporting every language, is bound to consume a significant amount of technical and expert resources within the tech companies themselves.

So, what is the point in developing this technology at all?

Where I really got to know the technology that sits behind these devices is while leading on a Conversational AI (Artificial Intelligence) initiative with HomeServe , delivered alongside Sabio Group using Google’s Dialogflow.

It was with this I could start to theorise the answer to my questions. Big data companies, such as Google, want to know how people transact, how they make requests, what dialects they use and what accents are available.

For example, UK English is not really UK English.

If you have ever sat around the table with someone from Birmingham, Newcastle, Glasgow, Cardiff, Bristol, and London, you will quickly find that there is no such thing as UK English.

Having a large volume of extremely specific repeatable transactions in all languages and dialects will turn the art of conversation into structured data which can then be analysed and modelled.

Within a home, the ‘home device tool’ sits there waiting to be called upon. However, within a contact centre, the tool is acting as the business’ virtual advisor for hundreds – if not thousands – of conversations on an hourly basis. It is asking each customer; “What is it you would like to talk about today?”, listening to the response and either continuing to fulfil the transaction or passing the customer an agent that could.

The financial model for these tech companies is to charge a fee for each question-and-answer response, which is significantly lower to host than a physical contact centre agent especially as you only pay for it when it’s working. If you can build a conversational model which is capable of handling part or all of a conversation, leaving only the complicated or high-risk conversations for the human, the business celebrates because it reduces their operating costs and supplies excellent insights into the workload within their contact centres. It also leads to a culture of wanting increased automation.

As the likes of Amazon, Google, Microsoft and IBM search for more conversations to be consumed in an automated way, the income per transaction also scales with it.

So, this may suggest that, although these technologies are currently making a loss as they scramble for market share in the contact centre, they may be in it for the long game.

In addition, and despite the digital revolution, nearly 70% of contact centre transactions are still conducted via the voice channel.

What this has done, is it has created an alternative look at the technology required within the contact centre industry, with self-service solutions now offered to customers in every channel where they can potentially transact.

Being the early leader in this market provides the largest opportunity to test and learn with not only enormous quantities of data, but also a range of differing approaches and methodologies in regards to implementation.

The most successful in this space, will potentially become the most lucrative in the long term – meaning today’s “colossal failure” might just be tomorrow’s “monumental winner”.

Phil Jordan will be appearing at Sabio’s digital transformation event, Disrupt, covering the potential of AI and Automation. You can find out more and register for the event here.

About the Author

Phil Jordan is AI & Automation Specialist at Sabio Group.

A key to retaining your existing business, growing your brand, and attracting new business is keeping your customers happy. Brian Tuite of Zenarate reveals new technologies that will help improve customer service.

HubSpot says 80 percent of customers reported that they have stopped doing business with a company due to a poor customer experience. Customer service agents are a company’s most important asset to keeping customers pleased.

Despite the rise of chatbots and tools to quickly solve customer issues, Harvard Business Review recently reported that 58 percent of customers prefer to call the contact center to solve urgent issues. The customer service role has evolved, but live agents will not go away. People want to talk to people – especially as automation and tools leave agents with the toughest problems to solve.

As agents continue to play a critical role in the customer experience, customer service departments need highly trained, confident and personable agents who understand industry best practices. The challenge for customer service leaders is to find the right method to train and upskill agents. Fortunately, new technologies help improve customer service agent performance before they even take their first call.

AI Conversation Simulation in Contact Centers

AI Conversation Simulation is a human-centered AI approach that provides realistic and impactful learning experiences through voice and chat scenarios. It creates hyper-realistic simulations of any voice or chat scenario, so customer service agents can learn through practicing, solving problems, making mistakes and building confidence. It’s like training in a live customer environment but without risks.

AI Conversation Simulation actively empowers agents to master high-impact customer scenarios quickly and conveniently to deliver consistent and superior customer experiences, leading to improved first contact resolution. Customer service departments build a culture of learning by enabling agents to practice, then share their best practice sessions with training managers and peers for open discussion and refinement of skills through human coaching.  Customer service agents report more confidence in talking to customers and a more enjoyable training experience with AI Conversation simulation.

Increased confidence translates to improved business metrics, like first call resolution, higher CSAT scores, faster speed to proficiency, and improved compliance scores. AI Conversation Simulation also improves retention and reduces hiring and training costs; because competent and confident agents stay longer, create a more positive environment and, most importantly, engage with customers better.

How to Create High Performing Agents with AI Conversation Simulation

Improving customer service agent performance requires a focus on helping people perform their best. With AI Conversation Simulation, this is done through an effective training approach that requires more doing and less facilitation. It also requires active listening and adjusting to customer needs, allowing agents to better relate with customers through enhanced soft skills.

At Zenarate, we employ some of the top minds in NLP and use the most advanced AI/ML to create life-like conversation simulations. We use advanced Natural Language Processing to allow agents to use their own words without scripting, and Natural Language Understanding generates real-time customer responses simulating a customer on any topic or persona. Our clients report exceptionally high user experience scores with agents citing how realistic and challenging their conversation simulations feel and how it improves their confidence.

We’ve created the #1 AI Conversation Simulation platform that supports global banks, large BPOs, and leading companies in the healthcare, technology, and services industries. With AI Conversation Simulation, contact center leaders develop top-performing customer-facing teams – delivering 56% faster speed to proficiency, 33% higher CSAT scores, and 32% lower employee attrition.

About the Author

Brian Tuite is co-founder and CEO of Zenarate, the leading conversation simulation solution for developing top-performing customer-facing teams. He’s held executive positions at MBNA, Bank of America, and SquareTwo Financial. Brian can be reached at briant@zenarate.com.

Demand for immediate, 24/7 support is at an all-time high. With the COVID-19 crisis causing a significant increase in customer inquiries, more businesses are turning to Conversational AI to sustain and improve customer and employee communications.

Conversational AI (aka intelligent chatbots or virtual agents) combines artificial intelligence (AI) and automation to streamline customer interactions across channels. Realizing the importance of transforming customer service or contact center operations and feeling the pressure to accelerate their digital transformation plans, many businesses have begun their automation journeys.

Yet, 95% are not successfully using AI, according to a CCW Digital Market Research study earlier this year.

What should business leaders take into consideration when evaluating Conversational AI solutions in order to increase their chances for success?

What to Look for in a Conversational AI platform:

Ease of Use – For Any User

Simplifying the creation and maintenance of virtual agents across the enterprise is critical. Enabling non-technical users (business users) to help build conversations, rather than only developers and conversational architects and designers, will add greater insights and perspective into end-user experiences. A broader team of users will also tackle the urgency of getting started, scaling and changing strategies faster.

A Platform to Handle Complexity

Customer service teams and contact centers have specific needs when considering new technology solutions. They need to support integrations with internal applications, session routing, speech-to-text and text-to-speech, and visual IVR. Selecting enterprise-grade Conversational AI platforms will meet these requirements, overcome legacy system challenges, and open new doors for future innovation.

Multi-Lingual Capabilities

In today’s global environment, it’s important to consider customer service tools that support bots in multiple languages. Bot platforms should have out-of-the-box Natural Language Processing (NLP) in enough languages and variants to support your customers’ needs. Automate translated conversations across channels, and incorporate fallback languages and local culture dynamics to communicate with your customers in the most natural way.

Accurate NLP

Natural Language Understanding (NLU) is a core component in any Conversational AI platform. Essentially, it is what makes a Virtual Agent smart. A powerful NLU engine achieves a high level of accuracy in understanding user inputs and knows exactly how to respond. A state-of-the-art NLP helps business users and developers stay focused on improving the NLU process and get significantly better results with less work.

Omni-channel and Voice Functionality

COVID-19 has created urgency in shortening the time-to-ROI for Contact Center Automation. Customers today expect a seamless omni-channel experience, whether on WhatsApp or a web chat or a phone call, and they want responses fast. An effective Conversational AI tool should plug into existing contact center solutions and enhance the standard capabilities with AI-driven technology. Particularly with voice channels, reducing call volumes into call centers, is a priority. Look for a turnkey, scalable solution for Contact Center Automation that allows customers to find answers to their questions before reaching a human agent.

Not all Conversational AI platforms are created equally. By aligning your existing requirements and wish list features with the right solution, your organization will not only improve customer service nd CX, but also accelerate time-to-ROI and pave new ways for future growth.

About the Author

Jessica Gopalakrishnan is the Senior Director of Marketing at Cognigy.

Cognigy is a global leader in conversational AI automation for contact centers. Its low-code platform, Cognigy.AI, enables enterprises to automate customer and employee communications using intelligent voice- and chatbots. Start a free trial of Cognigy.AI at cognigy.com.

The first-of-its-kind, proven framework shows businesses how to deliver modern, next-generation customer service with conversational support.

Intercom, the Conversational Relationship Platform that helps businesses drive faster growth, has announced the launch of the Conversational Support Funnel, the first holistic framework for delivering and scaling modern, next-generation customer support through a messenger-first experience.

The Conversational Support Funnel combines best-in-class proactive, self-serve and human support capabilities to get ahead of known customer problems before they arise, automatically answer repetitive questions with bots and quickly resolve complex issues with the right support agent.

According to PwC’s Future of CX Report, 59% of all consumers feel companies have lost touch with the human element of customer experience and nearly 80% of American consumers say that speed, convenience, knowledgeable help and friendly service are the most important elements of a positive customer experience. 77% of customers believe it takes too long to reach a live agent (Harris interactive / RightNow).

“Customer support is undergoing an irreversible transformation, and antiquated support systems just don’t cut it anymore,” said Jane Honey, Senior Director of Product, Intercom. Phone and email don’t meet the needs or expectations of modern customers and businesses have to find a new way to deliver customer support. Conversational support is the answer. Intercom’s business Messenger is the best way to build amazing relationships with customers because it’s personal, real-time, and fast. Businesses that don’t deliver conversational experiences will be left behind.

Businesses can’t afford to not deliver great experiences to their customers. According to PwC, 1 in 3 customers will leave a brand they love after just one bad experience, while 92% would completely abandon a company after two or three negative interactions. New Voice Media reports that U.S. companies lose more than $62 billion annually due to poor customer service.

Conversational support is the modern approach and the most effective way to meet customers where they are, with the personalized, convenient experience they expect, at scale. Intercom’s Conversational Support Funnel, developed after working with thousands of customer support leaders, shows businesses how to strategically scale conversational support beyond just live chat to improve the customer experience, increase business efficiency and boost the morale of customer support teams—effectively transforming these teams from cost centers to revenue-driving engines for businesses.

The three key support capabilities within Intercom’s Conversational Support Funnel include:

• Proactive support: Known customer questions can be answered proactively using tools like a business messenger for in-app and outbound messages or interactive product tours.
• Self-serve support: Common or repetitive customer questions can be answered and resolved automatically using chatbots and help center content, like knowledge bases.
• Human support: Complex or VIP customer questions are routed to human agents, who are equipped with management tools like an easy-to-use team inbox and powerful teammate workflows for greater efficiency and shorter reply times.

While many businesses are already doing parts of the Conversational Support Funnel, the framework connects these three pieces together for the first time. With this framework, businesses now have a proven approach to move beyond the customer support status quo and achieve better customer satisfaction and improved team efficiency and morale—all leading to stronger ROI.

Businesses looking to align with the Conversational Support Funnel will find that Intercom can do it all—real-time and asynchronous chat, conversational bots, contextual help content, targeted outbound messages, a team inbox with advanced rules and workflows for complex queries and more. No more cobbling together multiple tools to achieve what Intercom offers in a singular platform. Intercom can also be used by sales and marketing teams, so that conversational experiences at every part of the customer journey can be managed from one platform.

“We’ve seen companies who adopt conversational support successfully scale their support activities to meet and often exceed customer expectations—even during the COVID-19 pandemic when they faced an unprecedented volume of queries” said Jane Honey, Senior Director of Product, Intercom. “There’s no doubt that conversational support is the future, and Intercom is focused on empowering businesses with the very best tools and ways of working to evolve support to succeed in the new, modern era.”

Intercom’s Conversational Relationship Platform helps businesses build better relationships with their customers through messenger-based experiences and is used by sales, marketing and support teams to deliver conversational experiences across the customer journey.

Business demand for conversational support and strong upmarket growth is driving business growth for Intercom and the company surpassed $150 million in average recurring revenue (ARR) at the end of last year. Today, Intercom powers 500 million conversations a month and connects over 4 billion unique people worldwide across its more than 30,000 global customers, including Atlassian, New Relic, Shopify, Aer Lingus and Sotheby’s.

Find out more about the Conversational Support Funnel here.

About Intercom 

Intercom is a Conversational Relationship Platform (CRP) that helps businesses build better customer relationships through personalized, messenger-based experiences. It’s the only platform that delivers conversational experiences across the customer journey, with solutions for Conversational Marketing, Conversational Customer Engagement and Conversational Support.

Intercom is bringing a messenger-first experience to all business-to-customer communication, powering 500 million conversations per month and connecting 4 billion unique end users worldwide across its more than 30,000 paying customers, including Atlassian, Sotheby’s and New Relic. The company was founded in 2011 in San Francisco by Eoghan McCabe, Des Traynor, Ciaran Lee and David Barrett, with the mission to make internet business personal. Intercom was valued at $1.275 billion in 2018 and has secured $241 million in funding from investors including Kleiner Perkins, Bessemer Venture Partners and Social Capital.

Learn how a recent innovation called Conversational Process Automation is truly making it possible to understand customer intent, and do so cost-effectively.

We’re now living in an on-demand economy in which consumers insist on convenience, speed, and personalization. From ordering dinner with the push of a button to listening to a song by merely speaking its title, customer expectations are soaring higher and higher – and modern consumers aren’t inclined to wait for the companies they interact with to catch up.

For the customer service industry, this presents a substantial challenge. Not only are expectations of service rising, but demand within the contact center is as well. An increasing number of communication channels available through technology means a rapidly growing incoming ticket volume for customer service departments – one in which the cost of customer interactions is also ballooning.

With all of this increased demand, expectation, and cost falling upon the shoulders of customer service as an outcome of evolving technology, businesses are turning to technology for solutions of their own.

One of the key challenges that this tech will need to face is cost-effectively determining what the customer is hoping to accomplish when they reach out to the contact center. Whether by email, chat, web form, social media or otherwise, AI is already being adopted by many companies across the industry to support sustainable processes, but a recent innovation called Conversational Process Automation is truly making it possible to understand customer intent, and do so cost-effectively.

The Challenge

In order to serve the customer, the service department must first determine what the customer needs. The process to extract this information, when using traditional methods, is costly and can be inaccurate. It’s not financially feasible for business owners to hire and train enough human agents to answer and address each incoming customer call, so many have turned to other methods to try to remedy their situation.

For example, many companies use phone trees or web forms to narrow the customer’s intent. These types of tools help to reduce the possible scope of the need, but they are not an optimal solution. Many unique customer queries will not fit into a prescribed category, leaving the customer frustrated and the agent no better prepared to assist them.

It’s not only the dialogue where customer service reps need help. Our customers tell us that only 30 percent of customer service interactions in their contact centers involve actual “conversations.” When a consumer reaches out with an issue, chances are they aren’t just looking to chat about it, but for a transaction that provides them with a full solution to their issue.

Solving their issue requires the agent to take some type of action, often in one or more backend systems, which adds even more time and cost to the interaction. Customer service departments need to find a way to intelligently connect the conversations that determine customer need with the backend customer service processes that fulfill that need.

Conversational Process Automation provides that solution.

Unlike traditional customer service tools that work by deflecting a customer’s question, Conversational Process Automation (CPA) leverages machine learning to understand the request. For the most repetitive, time-consuming tickets, CPA can reach a point where it is able to determine the customer’s need, then resolve that need using all of the necessary systems without any agent involvement. This provides such a distinct advantage for both the customer and the agent because it doesn’t just delay a costly interaction, it can phase it out entirely.

With these monotonous tasks taken off of their plates, agents are able to tackle more complex issues with customers. While the CPA is tending to thousands of refund requests (or account lookups, or cancellations, etc.), agents can take more time on calls that require a more personal touch.

CPA is able to entirely take over these transactions because it employs open APIs that can interface with front-end CRM systems, as well as with backend software where ticket resolution actually occurs. Traditionally, customer service reps needed to switch between multiple programs in order to successfully resolve a ticket. They might have initially recorded information in their CRM system, yet when they needed to issue a refund or change a password, they were forced to pivot to another piece of software, and then back again. All of these extra steps can be avoided by streamlining the entire process with AI, with the results rapidly becoming evident in the bottom line.

The benefits of a collaborative relationship between the agent and the practical AI they’re using aren’t merely operational; it is also a great boost for morale. As they’ve taken on more challenging projects with more free time to be proactive, customer service reps are experiencing higher job satisfaction. Agents have self-reported higher employee satisfaction (ESAT) scores when using AI tools, which puts them at a lower risk of burnout, which in turn helps business owners manage turnover, decreasing recruiting and training expenses within the department.

With customer expectations and ticket volume on the rise, companies can’t afford to waste precious resources trying to extract customer intent. Businesses will increasingly turn to AI tools that utilize CPA to resolve the most repetitive and costly types of tickets they experience end-to-end.

As customer service becomes more accurate and timely, customer expectations of service will continue to rise, requiring companies to further step up their customer service operations. AI that gets smarter and learns over time is a crucial and necessary tool that facilitates their ability to do just that.

About the Author

Mikhail Naumov is the Co-founder & President of DigitalGenius, a venture-backed artificial intelligence company, transforming the customer service industry. In his role, Mikhail is focused on bringing practical applications of deep learning and artificial intelligence to customer service operations of growing companies and well-established enterprises.

As a frequent speaker on the topics of emerging technology, artificial intelligence & entrepreneurship he is a leading voice in the Human+AI movement, which focuses on the seamless interaction of human & machine intelligence in business applications and everyday life. Author of Amazon-Bestseller, “AI Is My Friend: A Practical Guide for Contact Centers” Recognized as Forbes’ 30 Under 30 for Enterprise Technology, Mikhail is passionate about bringing emerging technologies to life, to make business and everyday life more productive and enjoyable.