For quite some time we’ve become used to see live chat on many websites. It’s often used as the starting point for a customer service call, or to make a query about a new service we are interested in. With live chat we know there is a person (agent) on the other end answering our questions.
More recently the picture of the agent in the chat icon is often replaced with a ‘robot’ icon. This indicates we are no longer chatting to a human but rather a ‘chatbot’.
What is a chatbot?
A chatbot is basically a computer programme which is talking to us instead of a person. This computer programme can be quite simple and follow a heavily scripted conversation flow, or can be much more sophisticated and use artificial intelligence and machine learning techniques to interact with us.
Why are chatbots replacing real people?
The primary drivers are cost and efficiency. Chatbots don’t take holidays, don’t ask for pay rises and work 24 hours a day without breaks.
This can represent a huge potential saving if, for example, you run a call centre and many of the questions people call in with could be handled by a chatbot instead of an agent.
If the chatbot performs well you could even increase customer satisfaction (no more waiting on the end of a phone for an agent to become free).
If the software is clever enough it can appear to be very human like in the way it handles the conversation, however it can also get stuck and frustrate the user.
To be useful, chatbots need to be designed to have a very clear objective of what they are trying to achieve and to guide the conversation toward this objective.
Try and design a chatbot so it can go off topic, and you’re very likely to end up with the conversation in a dead end and having to revert to a real agent. If that happens too often, the benefits of chatbots are outweighed by the negatives.
Text based and speech based automated chat
Bear in mind also that chat can now be text based or speech based. More and more households now use voice based services like Alexa and Google Assistant. Talking rather then typing adds a whole new level of complexity but it can and is being done.
This leads us to the question of which service should you choose for your chatbots? There are many to choose from.
But some of the chatbot services from bigger companies like Amazon, Google and Microsoft have built in speech recognition and machine learning which reduce the development effort by decades to implement your chatbot. And make it massively cheaper.
UX and conversational marketing
But it isn’t just cost and efficiency. The chat ‘UX’ (user experience) is a whole new area of development which is looking at how we can use chatbots to better engage with clients and prospects.
Improved UX is driving innovation in ‘conversational marketing’ (CM). This allows us to engage with a prospective client at the point where we would have normally presented them with a form on a website and expect them to complete the form and then wait for us to contact them.
Chatbots allow a CM exchange to take place and better engage with the prospect at a crucial point in their product evaluation/buying cycle.
Many implementations of chatbots use familiar apps such as FB Messenger, KiK or Slack. This gives the user a familiar environment in which to have the dialogue.
Chatbots for business?
Should you be thinking about developing chatbots for your business? The answer is probably ‘yes’. Think about your business processes and which would benefit from chatbots and start a proof-of-concept. Set some goals and objectives for what you want it to achieve and would consider a success and then take action.
Developing a simple chat can be surprisingly cheap, with prices starting from just a few hundred Euros. Of course complex chat development can cost many thousands.
But you can start simple and then take it from there. You could well find that the business case is in favour of a sophisticated chatbot once you have worked out how you want to use it.