The use of chat interfaces for high volume interactions
BY JACK JOHNSON
As we look into the future we can relish the fact that emerging technology is helping reshape our future for the better. Technology is forever evolving and consistently moving towards cost effectiveness, ease, speed and sophistication at an extraordinary rate.
Sadly at the time of writing this blog insight, further cuts to local authorities were announced, with 36% of funding being cut by next year (2019) which is the biggest since 2010. Although ever increasing demand on public services and looming budget cuts threaten to stretch local councils even more, there are numerous digital transformation solutions to help alleviate building pressure. Admittedly the situation does not look optimal, however the ever increasing cost effective methods of dealing with customer enquiries provide hope.
What can help?
Hope manifests itself in the form of chatbots powered by AI (artificial intelligence). Chatbots are automated assistants that communicate through messages which simulate human interactions to help resolve customer inquiries. They can be integrated into a website, instant messaging or an app providing flexibility as to how they are utilised.
Cost of interaction
By employing such methods it will not only reduce costs and improve customer engagement but ultimately help relieve strain on services. Councils typically use traditional communication methods for their many county residents, who will contact them for numerous enquiries. But this comes at a cost. According to SOCITM, a face-to-face visit can cost an estimated £8.62 and telephone communication can cost £2.83 per interaction. However, online transactions only cost £0.15 meaning they are 18 times cheaper than phoning. These results were published in 2011 and the landscape of these costs have shifted by looking at Ealing Council calculations published in 2017 which saw the telephone cost drop to £0.41 by 2014/15.
Council case study of self-service
Blackpool Council produced a case study detailing how they integrated a self-service system that could be accessed 24/7 to help resolve waste collection requests. The system in question was published in May 2015 and by the end of March 2016 saw 1192 residents switch to the self-service method saving an estimated £309 (based on £0.41 a call). However, although this case study seems to provide impressive results in a time where saving money is critical, there is no way to measure the customer experience.
Use cases for chatbots
Companies boasting vast customer bases dealing with a variety of queries on a daily basis have utilised chatbots to help with the sheer volume of interactions. Companies such as Spotify, Lyft, Starbucks, Duolingo, Whole Foods and many more have implemented chatbots in numerous ways in multiple fields to help deliver their services at a faster rate. 69% of consumers prefer chatbots for quick communication with brands. The perfect implementation seems to be for handling simplistic requests from customers which inevitably helps reduce interaction times for consumers.
Companies like Lyft have leveraged the significant use of Facebook to integrate their chatbot with Facebook Messenger to help extend their service reach. It allows customers to request a ride and the bot will keep you updated on the driver location and show the user an image of the car and license plate. This service is also available through Slack allowing for more accessibility. This example workflow is very encompassing of other company examples showing how most chatbots are used to fulfill customer requests and enquiries.
There are 2 types of chatbot which differ dramatically. One which acts within a set of rules requiring a smaller skillset and is limited to answering specific questions meaning limited usability. The other type is machine learning which learns from each exchange and actively seeks context to help determine if an answer will resolve the customer question.
Royal Bank of Scotland have adopted a chatbot titled Cora who has 1,000 responses to over 200 customer queries. The chatbot will map out all possible replies to a customer query when received and estimate how likely each one is to answer the customers message. If the confidence of that answer is above a predetermined threshold it will reply automatically. If the level drops below that threshold or they reply no to “Has this answered your question” the chatbot will typically invite a customer service employee into the chat to help resolve the problem. This provides a very unique workflow allowing for employees to spend less time dealing with generic enquiries.
There are many more examples of companies using chatbots to help deal with these high volume interactions. By reviewing these cases we are able to see that the implementation of chatbots have been highly successful and have in fact managed to increase customer engagement.
Informal versus formal
It is also crucial to consider the tone of voice a chatbot should project to the consumer. 24% of people would stop using chatbots if it wasn’t able to chat in a friendly manner. Although consumers know they are chatting to a bot that 24% need to be considered and that there is an empathetic more human side to it. However, the tone of voice is dependant on the brand and the service being provided. CNN provide news via their chatbot and as such should provide a serious tone to ensure informal text does not accompany sensitive news articles. However, CNN opt for a lighter tone when people are first signing up for the news updates even including a GIF. In cases where the bot can afford to be informal there still needs to be great regard for the choice of words used to ensure they resonate with consumers better.
Customer satisfaction should be a digital priority for companies, with 73% of customers listing it as an important factor in their purchasing decisions. The ability to monitor chatbot analytics companies are better able to equip the AI to deal with requests in a more streamlined way. The availability of the service staff still allow consumers to chat with a person if they wish. According to Drift, 64% of those asked believed that a 24 hour chatbot service would be a benefit, 55% believed an instant response would be, with 55% believing that simple answer questions were also a top benefit. What is positive about this data is that chatbots are already supplying these services to consumers.
Chatbots are poised to become the face of customer service in the near future by providing real-time information. By using chatbots alongside online chat it really seems to be the ideal solution for businesses to help deliver a faster online experience for for their consumers.
We help teams across the UK deliver successful digital projects. If you’d like to have a conversation with us, we’ll put the kettle on.