The Importance of Chatbots: The Online Virtual Assistant

Chatbots:

The one of the main initiatives that stood out from the strategy proposal that can help a business build their website, whilst improving their online engagement are chatbots. The concept behind chatbots are that they are online intellectual virtual assistants, which can achieve a conversation with customer’s online, using natural language whilst in a written form (Dale, 2016).

The forefront of chatbots derived from companies such as Apple, Microsoft, Google and Amazon; who all have their own forms of chatbots, including ‘Siri’ from Apple and ‘Alexa’ by Amazon. Many huge businesses can see huge potential in this digital technology, as it supports companies trying to interact with huge volumes of customer queries and complaints.   Shriftman (2017) mentioned how chatbots help customer experiences and goes beyond the simple shopping process of peruse and purchase. In many ways, it becomes a digital expert for the items customers are looking for, each specific to their websites.

 

This blog will now further look into how chatbots can be implemented successfully, alongside the risks associated with them.

 

How can a Chatbot system be executed well? How can we make one?

Wilcox & Wilcox (2013) have described how the best chatbots made have been ones that have fooled actual judges in competitions that they are a real human. They also went onto detail that the best chatbots on a global scale are able to tell a life story, whilst maintaining an unfailing personality and providing emotional responses.

 

According to Wilcox and Wilcox (2013) there are 8 main factors to consider when creating a chatbot:

  1. ChatScript: The first main factor is a ChatScript, which is the foundation process for a chatbot. It is noted as an expert system which matches inputs of patterned meanings towards specific outputs. Within a chatscript, there are many factors to be taken into consideration. These include rules, pattern matching, standardization, macros, concepts, topics and parsing and introspection. By this point, a basis for the chatbot has been created.

 

  1. Personality Design: The second stage to form a chatbot is personality design; this stage involves writing a biography to give the chatbot a character to make it more realistic to the customer.

 

  1. Designing a Topic: The third factor is designing a topic; this stage converts the biography and from third person to first in order to create a human effect.

 

  1. Writing Script: The fourth stage involves writing the full script. This process involves programming all the potential questions the chatbot could get asked along with the suggested answers for them. Additionally the chatbot could also get programmed to ask questions and its potential responses from the user. This is very complex and time consuming, but the more in-depth the programming is, the more personal the chatbot will be with the customer.

 

  1. Testing: The three testing methods that are suggested by Wilcox and Wilcox (2013) are verify keywords, verify patterns and verify blocking. These all make sure that there are no anomalies in the chat script and confirm whether it’s ready to be implemented on the website.

 

  1. Integration: This stage is when the chatbot is installed on the website and ready for use to the public.

 

  1. Evolution of the process: Following integration, the chatbot needs to be regularly checked to see if it is satisfying the customer and working to its best potential. This can be analysed through statistics which show the number of chatbot responses and how successful they were, along with exactly how the customer wrote to the chatbot as well, giving the programmer areas for potential improvement.

 

  1. Valley of Chat: After observation of the chatbot over time, Wilcox and Wilcox (2013) mentioned how people can abuse the chatbot with the random questions they ask for fun. They indicate that at this stage, it is up to the creator whether they want to cater to this nature of chat or not.

 

 

What are the risks and important information associated with Chatbot systems?

Although chatbots are a great tool to improve relationships between companies and their customers, they do have their challenges. The system behind making a chatbot is very complex, needing great care and attention in order to give the customer the correct information.  Thorne (2017) provided alternative, broken down methodology for creating a chatbot and indicated that there are three key components involved in making a chatbot; they include:

 

Thorne (2017) mentioned how creating a chatbot is time-consuming and could even cause further frustration for the customer and that even this three stage process can take up a lot of time and resources, even potentially from other areas of a business, especially if they are small. Thorne also indicated that chatbots can also only go so far with conversation,

meaning that they can only engage in very uncomplicated and short conversations. From this, in general businesses would need to take into consideration if they really need a chatbot and why so. It is only recommended that huge companies with huge lines of products/services acquire a chatbot, as they have the time and money in comparison to small businesses and it would serve more beneficial to them.

Furthermore, Dhanashetti (2017) discussed how chatbots have a high error rate and are limited in regards to their capabilities. As much as we would like to make them into a human like system, it will never be sophisticated enough to discuss everything in-depth. Dhanashetti (2017) even went as far to state that business should still stick to a human-led engagement strategy with consumers and to not over-rely on chatbots.

 

Feel free to check out these links below to further information on Chatbots:

YouTube link: https://www.youtube.com/watch?v=JGIFN9HHl04

Article (Beginners Guide)https://chatbotsmagazine.com/the-complete-beginner-s-guide-to-chatbots-8280b7b906ca

 

References:

Dale, R. (2016) “Industry Watch: The Return of the chatbots”, Natural Language Engineering,Vol. 22 No. 5, pp. 811-817.

 

Shriftman, J. (2017) “5 Examples of Luxury Fashion Retailers Using Chatbots” Snaps [online] Available at: https://makesnaps.com/2017/12/11/5-examples-luxury-fashion-retailers-using-chatbots/  [Accessed 20/02/2018].

 

Thorne, C. (2017) “Chatbots for Troubleshooting: A survey”, Language & Linguistics Compass, Vol. 11, No. 10, pp. 1-14.

 

Wilox, B. & Wilcox, S. (2013) “Making it Real: Loebner-Winning Chatbot Design”, Arbor, Vol. 189, No. 764, pp. 86.

 

Dhanashetti, N. (2017) “What’s good and bad about chatbots?” Quora [online] Available at: https://www.quora.com/Whats-good-and-bad-about-chatbots [Accessed 19/02/2018].