Chatbot, RPA and artificial intelligence: trio of true process automation


chatbot, rpa and artificial intelligence: trio of true process automation

To be successful in automation processes, it is essential to cover the entire end-to-end cycle. From people (clients, suppliers, workers, etc.) until the last process included in a certain workflow. In this article we will discuss how, thanks to the combination of three different but complementary technologies, it is possible to achieve this by guaranteed quality process and providing users with an optimal experience.

If we draw the different layers that intervene throughout a process, from interaction to optimization in SAP, we observe three complementary technologies:

  • Conversational Artificial Intelligence: It is the interaction tool, the interface with people through what we know as a chatbot, a software that manages the conversation in order to obtain the necessary information to execute the process successfully.
  • Robotic Process Automation: It is the execution tool, which carries out the repetitive tasks in an automated way, either through APIs, services or the simulation of screens (from SAP, other programs or web pages ), filling in fields and pressing buttons.
  • Machine Learning (ML): They are the set of algorithms that provide the intelligence necessary to optimize the data and generate conclusions as similar as possible to what the human interaction would process.

These three components maintain a continuous interaction and a constant exchange of information, in the same way that a person would involved in the development of the process. For example, to provide the appropriate responses during the conversation it is usually necessary to have information from various programs (RPA), or to activate different machine learning algorithms to better understand what the user is saying and offer accurate and quality responses.

Likewise, during the execution with RPA we may need more information from the user, which leads to relaunching the conversation (CAI) or resorting to the activation of ML algorithms to infer information based on how the process has been solved previously or through a calculation of probabilities.

In short, it is a process of continuous interaction, execution and optimization until it is concluded without errors and with the user's full satisfaction. Eventually, the maturity market is reaching, becomes a compelling reason to adopt this suite of technologies.


Seidor we are verifying that the development of individual projects is no longer an option. Success in this type of initiative depends on the adoption of a measured and controlled strategic approach that allows scalability through languages, channels and the company itself, and for this, the power and versatility offered by the latest technologies is essential.

Regarding future prospects, it is expected that, in the coming years, clients will be able to manage most of their relationship with a company without requiring interaction with a human. In this paradigm, conversational intelligence, even activated by voice (we are talking about the integration of devices like Alexa at the company level), will play a fundamental role. However, there are still many challenges to be solved and at 
Seidor LAB we have been working on them for some time.

"It is essential to apply a layer of global vision and digital transformation."

In parallel, the integration through an intranet of conversational AI applications is positioned as another of the key evolutions. This intelligent routing will allow the inter-application handover process to take place in a number of ways. These include the possibility of having a master application or a superbot capable of guiding and delivering various processes in a fully integrated way. For this, it is essential to apply a layer of global vision and digital transformation and at Seidor we deploy the appropriate strategies for its correct implementation.


All this analysis work has allowed us to identify the main challenges for each of the technologies:

  • Improve the chatbot customer experience, since they not only have to answer their questions, but they should also speak, think and, above all, develop emotional relationships with users, carrying out an adequate analysis of feelings.
  • Apply the latest advances in natural language processing and unsupervised learning and reinforced learning algorithms that provide chatbots with sophisticated algorithms that allow them to provide more unique and personalized experiences, creating more authentic relationships with a specific target audience.
  • Shorten the time required to have the bot prepared with sufficient quality without the need for long periods of simulation. For this, previously trained models of various types are used.
  • Provide accelerators to the processes most likely to be automated in the SAP product ecosystem, both from the point of view of automation in RPA and the intelligence applied for optimization.
  • Integrate other technologies that can enrich automated processes, such as blockchain for processes that involve customers, consumers or suppliers.

Finally, keep in mind that, in certain moments of interaction, sometimes there is no substitute for the empathy that human agents can offer or the kind of intelligence that creativity needs to solve a case. In these situations, it is often the human ability to draw parallels with similar experiences that enables problem solving in complex or unusual circumstances.

Ultimately, it is essential for a chatbot to be able to seamlessly interact with a human agent when necessary and ensure that all the information collected is transferred so that the customer does not have to start over with the frustration.

New call-to-action