5 essential considerations for chatbots in the Enterprise – Part one

Enterprises are slowly waking up to chatbots. Chatbots are going to be crucial tomorrow in engaging various constituencies, candidates and hires, employees, vendors and consultants.


Chatbots are here to stay with initial applications in the consumer space with offerings from Apple, Microsoft, Amazon and google going mainstream. The popularity of messaging platforms has been a steeper adoption curve compared to the internet, mobile and social media and the use of chat applications like FB messenger, whatsapp, Viber, Kik, Telegram, Slack, Skype and others exceeding the use of social media.


3 of the top 10 app installs are chatbot applications which are increasing in use daily with expectations of 2.5 billion active daily users on chat projected to reach soon.


Chatbots leverage the familiar conversation interface and piggyback on messenger and chat platforms thereby circumventing a problem faced by most mobile apps, that of app uninstalls. Our mobile phones now routinely sacrifice apps first to store entertainment media, but we never uninstall crucial chat based applications. On average customers use chat applications 23 minutes a day globally and expected to grow to 50 minutes on average globally.


Enterprises are catching on to this opportunity. A chatbot is the most natural way for enterprises to engage with all the people that interact with the brand. A conversational medium like a chatbot reduces the time from thought to action. Queries, performing transactions and getting status on service, billing, payments and a whole slew of business transactions is the fastest via a conversational media like a chatbot. It obviates a need to follow a map (menu), a structure, difficult to find layered information and hidden specifics.


Organizations are realizing the benefits of employing chatbots to boost productivity, provide better service and create better engagement with employees and customers.


Chatbot applications can be primarily classified as Internal or External. Internal Chabot apps bring about efficiency in Human resources, help desk and service desk tickets, policy, compliance documents search and various other use cases specific to employees.


External apps are useful for sales and marketing, business transactions, customer service and support, supply chain and delivery, hiring and scheduling and a whole slew of use cases that an organization needs to perform on a day to day basis. With systems now being relatively real-time across functions that enterprises have heavily spent on in the last 2 decades and the prevalence of the internet and ease of access via smartphones, we must now focus on ease and speed of query addressal, faster transactions throughputs, more pinpoint queries  and move away from clunky and dated web and mobile interfaces that force a path on the user.


So, what are the requirements for employing a chatbot in an enterprise?



Most enterprise users use sophisticated applications having a rich user interface. Asking them to switch to only text messaging or voice may not find many takers. Using visual elements like menus, buttons, scroll bars mixed with text and voice will be more acceptable. Having a flexible and rich Chatbot UX will be a key criterion in increasing adoption.



This will perhaps be one of the major stumbling blocks for the adoption of chatbots in an enterprise like the resistance when mobility made its entry into the enterprise. Most chatbot platforms operate in the cloud and will have to access information and interact with entities inside of an enterprise and outside too. Most enterprises have strict security compliance rules to adhere to and chatbots will be expected to be in the clear. Strict access control mechanisms, data encryption – in motion and at rest, data protection – in motion and at rest, data retention, authentication – multi-factor, authorization and multi-tenant policies have to be set up. Enterprises will expect the chatbot platforms to comply with the security standards and be up to date for handling new vulnerabilities.



As mentioned above most chatbot platforms operate in a public cloud. And chatbots are expected to interact with users inside and outside of an enterprise. They also have to access information stored inside an enterprise and sometimes from external public services. Integration will be a challenge and any platform that can provide seamless integration capabilities will have a significant advantage. Rest API integration and webhooks will be the default choice. However not all enterprise will provide or allow API integration to its internal systems. In such cases providing an enterprise gateway software to integrate with the internal systems on one hand and the cloud-based platform, on the other hand, would be an option. Custom integration software may have to be written for legacy systems and plugged into the Gateway as an add-on module.



Large and global enterprises will process enormous chatbot interactions per second. Chatbots will be expected to horizontally scale to handle thousands of simultaneous users and tens of thousands of interactions per second with ease. Additionally, availability will also be key criteria and the chatbots and the chatbot platforms will have to provide 99.99% uptime or more since they will become business critical entities for an enterprise.



Enterprise Chatbots will exhibit the characteristics of a big data platform i.e. volume, velocity and variety. Under such circumstances, performance of the chatbot interactions and the chatbot platform will become a major factor for surviving in an enterprise. In today’s business, a delay of milliseconds can cost a huge amount of money especially for the chatbots operating in fin-tech area. Since chatbots have to access internal systems in the enterprises, any slowness caused by them will first reflect on the chatbot’s performance.

In part two of this article, we will cover Customization, Natural language understanding (NLU), Domain knowledge, Auditing and Internationalization requirements for enterprise chatbots.



By Anwesh Roy
Engati – www.engati.com – A Coviam technologies platform
Vice President
Coviam Technologies