There has been a lot of hype around chatbots and virtual assistants, but looking past all that, how do you establish if you really do need a chatbot or virtual assistant?
If your business is looking at making that decision right now, you may feel caught between not wanting to rush due to the hype about conversational AI and the fear of losing out to your competition who are giving customers a better experience than you, and might be gaining an advantage.
Chatbots as well known and there are both good and bad examples of the technology. Early examples followed basic rules and workflows, were clunky and offered very little value. Recent examples can be both intelligent (through conversational AI and emerging technology) and connect to all sorts of data to add powerful functionality and compelling results. These are more often referred to as Intelligent Virtual Assistants or Virtual Agents. These can understand a wide variety of customer enquiries whilst delivering the responses through eloquent and natural sounding speech, through messenger or even digitally face to face as if they were a digital human.
By integrating your virtual agents to your business systems you can replicate what your customer service team do. Intelligent virtual agents have a conversation with your customer, follow processes that verify the user, look-up customer information, intelligently respond and update systems, triggering further actions or workflows.
Customer service chatbots are great at reducing costs, but the improvements they make in customer experience can have even more impact. Available 24 hours a day, 7 days a week, and able to answer customers’ questions instantly whilst intelligently handing-off to live advisors when necessary, properly designed virtual agents can vastly improve customer satisfaction.
When looking at implementation of virtual agents, businesses should consider the profile of your audience and when it is right to automate conversations vs use live advisors. With the right platform you can triage all initial enquiries through the phone, chat or video, and follow workflows that either handle the complete enquiry or hand over to your existing customer service offering with the full call history.
Is a Customer Service Chatbot right for you?
Virtual agents are most effective in customer facing sectors where there are frequent and monotonous requests, but anywhere that has high levels of customer interaction can benefit.
Previously it was thought that deployments would offer most benefit to companies fielding thousands of customer chats or calls via contact centres with hundreds of agents. This was due to the amount of data needed to train the AI, the length of development time and the cost to run.
However, as chatbots and virtual assistants have evolved, SaaS based pay-as-you-consume models were introduced and sector accelerator packs with existing workflows built, it has become much more accessible to organisations with smaller requirements - and budgets! The development of low-code and no-code design has meant that workflows can be created in internally without the need for a specialist trained IT department.
In fact, virtual agents can even help those without a contact centre at all - see our blog Can I Use AI if I Don't Have A Contact Center?
Can chatbots be deployed in areas other than customer services, like sales?
The best results are currently being achieved in customer service environments, but even here they are also being used to cross-sell or up-sell to existing customers. In this scenario the chatbot can use available data or understand key trigger words or phrases to change its behaviour and responses. This could be enquiries about a high value product where you want to transfer to a specialist advisor or identifying a need where the workflow must change.
They can also be used for internal customers such as onboarding for HR, or support for the IT Helpdesk, but they are also starting to show great results in sales and marketing environments. Find out more in our blog Is AI Just for Customer Services?
Choice of Platforms
Some of the more basic customer service applications can use the low cost or open source platforms. However, authentication, privacy and data security provide challenges to many looking to implement a chatbot, and many of these lower cost platforms may use your data to develop their learning models.
The ideal platform will allow you to use the AI technology from the leading global providers and utilise the services best suited to your requirement. No one knows which vendors will lead the way in 1 years time let alone 5 years from now but there are already substantial differences in capabilities between the providers today.
If you want to deliver a simple messaging based chatbot then almost any platform can offer a reasonable solution. If you require a speech based service that integrates with your telephony system and contact centre, collects data from your back-end systems using precision speech and reformats the information to make it conversational friendly then you have very few choices.
When you also want to provide a platform where anyone can build complex workflows and integrate to knowledgebases through low-code/no-code portals and use templates to rapidly get up and running the options become even more limited.
Whilst it is great to have big ideas, one of the benefits of the new SaaS based pay-as-you-consume model is that you can start with small pilot projects where you can demonstrate success. After the system has proved itself, you can expand into other areas.
One of the major advantages of conversational AI-based systems is that they can either self learn, or suggest training phrases for you to approve and ensure it improves over time. A virtual agent will just keep getting better!
The decision to deploy virtual agents can be complex, but it is the way most customer services organisations are going because it’s faster, and better, for the customer.
If you need help in planning and implementing you customer automation journey contact us or call us on 0333 6000 360.