<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=283259072089923&amp;ev=PageView&amp;noscript=1">
Get Started for Free

The 3 Evolutions of Bots for IT Service Desk

Mar 30, 2018 1:06:51 PM


The Modern Era of Intelligent Enterprise Automation

With the rise of chatbots and AI, enterprise tasks and processes which were previously difficult to automate are now experiencing rapid cost and labor efficiencies.

This is largely due to the latest wave of automation taking place within Human Resources (HR), Customer Support, and IT Service Management (ITSM).

In this article, we will discuss the three evolutions of bots for IT Service Desk automation in order of capability from basic to most advanced.

The Automation Spectrum for
IT Service Desk Bots

Timeline@2x-100 - edited

Evolution One - Robotic Process Automation (RPA)



The first evolution of IT Service Desk automation came in the form of robotic process automation (RPA). RPA helped enterprises transition from providing labor services to software-as-a-service.

Repetitive tasks that used to take employees significant time to complete, could now be automated using RPA bots which would mimic a series of human clicks/interactions and put repetitive tasks on cruise control.

IT Process Automation (ITPA) is a form of RPA which is focused specifically on the needs and pain points of an enterprise’s IT infrastructure. Since IT departments rely on multiple IT systems, processes, and workflows, ITPA requires a more intricate level of programming in order to handle more complex tasks (i.e. multi-factor authorization and password resets, virtual server provisioning and configuration, cyber security incidents,  etc).

For Example:
In the case of a password reset, IT service agents would no longer need to login to an application and manually request a password reset. Instead, the agent could utilize ITPA to fully automate the login and password reset process on the backend.

As a result, ITPA gave rise to a virtual workforce which helped enterprises reduce unnecessary human effort and dramatically improve operational efficiency.

Challenges with RPA/ITPA

The biggest drawback with RPA/ITPA is that individual task automation has lower value and a narrow scope whereas automating an entire IT process has higher value and broader scope. This leads us to our first big question:


How do enterprises apply the automation capabilities of RPA/ITPA towards entire IT processes, rather than simple tasks?


This brings us to our second evolution of IT Service Desk automation.


Evolution Two - Intelligent Automation


The second evolution of IT Service Desk automation is bots powered by natural language understanding (NLU).

While ITPA bots can handle repetitive IT tasks with ease, they often stumble and ultimately fail when encountering more complex requests in unstructured formats.

NLU-powered bots maintain all the benefits of RPA/ITPA but are able to understand complex user requests and intelligently automate processes based on the understanding of those requests.



Due to major advances in NLU capabilities over the past decade, enterprises have embraced intelligent automation and have started deploying chatbots which understand human input and connect with ITPA bots to execute tasks.

This results in IT Service Desk bots that can automate the end-to-end interactions between employees/customers and the enterprise.

Challenges with Intelligent Automation

The challenges with intelligent automation revolve around the scalability and consistent accuracy of the IT Service Desk bots over time. Creating, training, and maintaining a bot’s skills is currently a very manual process that does not scale at volume.


Creating and Maintaining an IT Service Desk Bot (Manually)


Here is the 4-step process of creating and maintaining a bot:

  1. Manually identify the service requests that need to be automated
  2. Manually train an NLU model to enable the chatbot’s understanding of each request
  3. Manually build a bot and test it for consistent accuracy and performance
  4. Once in production, repeat steps 1-3 using chat logs to add new skills to the chatbot

Many enterprises have piloted chatbots in narrow areas and are seeing great success, but the number one question they have is - how does this scale?

This leads us to our next big question:


Once you build, train, and deploy an IT Service Desk bot, how do you sustainably scale its capabilities as your enterprise evolves?


This brings us to our final evolution of IT Service Desk automation.


Evolution Three - Intelligent Automation at Scale

The third and most advanced evolution of IT Service Desk automation involves bots that use NLU and can leverage existing enterprise data to automatically maintain and improve their capabilities.

Large enterprises have thousands of recorded employee and customer interactions that are archived in their CRM, Service Desk, or Customer Support systems.

What if you could harness this data to create an out-of-the-box AI-based chatbot customized to your enterprise?

Now that would be super cool! Wouldn’t it?



Well, we have fantastic news. Using existing IT service tickets, enterprises can now use unsupervised learning to automatically extract intents and entities from their tickets, surface the most frequent intents and entities using clustering techniques, and fully automate them.


Creating and Maintaining a Service Desk Bot (Automatically)



Challenges with Intelligent Automation at Scale

Unfortunately, if you want to build these robust and scalable IT Service Desk chatbots in-house, you will need to spend significant engineering resources, utilize immense computing power, and tweak core NLU algorithms on an ongoing basis.

As we discussed in our previous blog post, “The 3 Essentials of Building AI Bots for Enterprise IT Help Desk”, it is critical for enterprises to quickly embrace AI-powered IT Service Desk bots and intelligent cognitive automation. However, the implementation must be done correctly in order to see a quicker resolution of IT service requests and significantly reduced operational costs.This is where Parlo can help.


Introducing Parlo IT Service Desk Bots


At Parlo we believe in democratizing the process of creating intelligent automation at scale by automatically detecting intents from your tickets as a service.

Over the past 2 months, we ran a closed beta with a few customers and the results have been promising. Our beta customers have been able to streamline the process of creating, training, and maintaining a bot for IT service Desk requests. 


Right now we are offering a limited number of enterprises to create, train, and maintain IT Service Desk bots out-of-the-box. If you are interested in this solution for your enterprise, please submit a request to our team.



We’ve come a long way from the days of using human agents to handle routine IT service requests within an enterprise. Here are the three evolutions of IT Service Desk bots being used within enterprises today:

  1. Evolution One - Robotic Process Automation (RPA)

    • RPA bots help enterprises automate routine tasks in order to reduce unnecessary human intervention and improve operational efficiency.
  2. Evolution Two - Intelligent Automation

    • Intelligent automation involves NLU-powered bots that can understand complex user requests and intelligently automate processes based on the understanding of those requests.
  3. Evolution Three - Intelligent Automation at Scale

    • Instead of requiring manual training and maintenance, these NLU-powered bots can leverage existing enterprise data to automatically maintain and improve its capabilities.

The future evolutions of IT Service Desk automation will involve end-to-end deep learning. However, given our current state of AI technology, we are at least 2-3 years from this becoming a reality. Until then, Intelligent Automation at Scale is the most pragmatic approach to achieving IT Service Desk automation.

We are proud to announce that Parlo’s latest IT Service Desk cloud offering is available for customers to use in open beta now! So long as you have CRM or Service Desk data, you should be able to create a sophisticated chatbot in 60 days or less.

So why wait? Check out Parlo’s Open Beta program to learn more.

Get Started for Free


If you found this article useful, please let us know your thoughts below. We regularly publish content like this on our blog in our mission to bring smarter AI to smarter enterprises.

Parlo builds smarter bots for smarter enterprises. We build AI chatbots that employ cutting-edge machine learning to seamlessly integrate with your business. They’ll support your human workforce, delight your customers, and save you time and money.