ISG Provider Lens™ Intelligent Automation - Platforms and Products - Process Discovery and Mining - U.S. 2022
Various RPA solutions are converging into a platform.
Intelligent automation is now a mainstream technology spanning across enterprise portfolios. The evolution of RPA platforms in the past decade has been astounding, and the technology is now a vital constituent for driving digital business transformation. Core RPA characteristics are being integrated into platforms for building advanced products and capabilities. These platforms converge proprietary capabilities, emerging technologies and third-party solutions to develop a unique value proposition. This continuously evolving category acts as a differentiator and is one of the significant developments in the software vendor market.
With the increase in digitalization and end-user expectations, enterprises are compelled to provide on-time resolution for service continuity. Technology teams are building AI capabilities and components to improve communication quality and effectiveness across all interaction points. However, deploying flow-based conversational assistants is complex and perplexing due to the multilayer data fabric and intelligence models. Business transaction, user and system data are leveraged to address industry and function-specific queries. The technology teams working on designing intelligent automation solutions are exploring new and unconventional approaches. One such example is learning from continuous streaming of audio-visual data generated by regional sources to understand the dialects of a language for improving accuracy and relevance. Domain ontologies are being assembled for specific industries’ verticals to enable conversational engines and simultaneously support business users and customers to enhance engagement and efficiency. The aim is to conduct human-like interactions to address and predict customer wearies and requests.
Conversational AI
Strong year-over-year growth for conversational AI solutions is expected to continue. North America is projected to have the largest market size owing to the increasing demand for enhancing customer experience and retention initiatives among enterprises in the region. Technology giants in the U.S. are acquiring AI capabilities and product companies to strengthen their offerings portfolio. The U.S. has been at the forefront of technology, being a leader in applications for AI patents.
The most common and preferred enterprise applications of AI is for conversational interface or bots. Chatbots increasingly use text and voice to communicate via business apps. However, conversations are getting complex, multi-layered and flow-based with the improvement task open for all functional and technical teams. To address this, conversational structures are used to design interactional experiences between users and businesses to ensure prompt and accurate information provisioning. These conversational interfaces can make recommendations based on past preferences, user choices, records and interactions. Conversational platform and product companies are ambitiously working towards incorporating voice and text capabilities, enabling stronger interactions and greater engagement across all stakeholder segments.
This typically involves integrating machine learning, natural language processing (NLP) and speech-based technology into a single platform that engineers can use to develop and build applications across various business functions, adding the flavor of proprietary components.
The convergence of conversational AI and intelligent automation has created a new stream of applications across different industries to create an unmatched experience for businesses and enterprises. Furthermore, it enables users to interact with technology, leveraging NLP to make cognitive business decisions. However, the use of only conversational capability solves a limited set of use cases. When complemented by automation, the value increases multifold and improvises business outcomes.
Intelligent document processing (IDP)
Digitalization of business information and conversion of transactional knowledge into a digital format is not only the priority but a step towards environmental sensitivity. The conversion of analog data formats to machine-readable and consumable structures improves the productivity, accuracy and archiving of information, elevating the concept of information on demand.
IDP involves analyzing and processing unstructured and structured data to drive actionable insights and provide end-toend automation to the document-centric business process. In addition, data can be documented in various forms, including paper documents, faxes, emails, PDFs, attachments and Microsoft Office files and stored in different locations (on-premises or cloud).
Therefore, it is vital that all documents are processed intelligently so that business information can benefit the organization by cost effectively and efficiently integrating IDP into its workflow.
IDP adoption is increasing at an equal pace as other AI and machine learningbased technologies. Nevertheless, the IDP platform market will be valued above $5 billion by 2023 according to one market study. The primary reason for the cautious adoption is the lack of awareness of such platforms and their capabilities. Enterprises often misapprehend IDP platforms and their benefits to document capturing, indexing and labeling, but the capability extends to extraction, parsing and restructuring. Additionally, the everchanging compliance and regulatory environment is a significant factor for its calibrated growth.
IDP components are crucial to transform paper and digital documents into structured extracted data at every step of processing. Document capture is an integral part of the platform as it ingests raw data by scanning the entire document. The IDP platform utilizes computer vision and image processing techniques to identify and extract data from various document types. Integrating several AI capabilities and automation has several applications in the real world across different industries. IDP vendors have built industry vertical products and kits to resolve perennial challenges in handling documents.
Process discovery and mining (PD&M)
Enterprise process optimization focuses on refining processes and attributes, as it is necessary to identify areas where the automation and reengineering of tasks are essential to avert effort, efficiency and accuracy leakages. In addition, processes are the carriers of business information that change with methods, modes and mediums of interactions. Therefore, any inefficiency in the process directly impacts business outcomes and customer experience.
Process discovery and task mining are synonymous terms that focus on capturing the extrinsic actions and activities executed for the input or output of a task. These are a combination of actions and information through a system interface to complete a transaction. Process mining extends the capability of identifying, analyzing and establishing a pattern in the intrinsic system and user data through machine learning and deep learning algorithms. A set of algorithms constantly analyzes and learn from the patterns to present an inference to simplify, optimize and enhance the process outcomes.
Business process management and advanced capabilities are increasingly gaining prominence in the process discovery and mining space. These capabilities add value, improvise how processes are executed in an environment, and capture variations to identify the underlying pattern. Predictive analytics has played a significant role in flagging deviations. However, prescriptive analytics is a beneficial and niche area that helps with calculated and calibrated directions.
Automation platforms converging with AI and machine learning technologies can simplify business processes across all corporate functions to improve the overall business efficiency. In addition, reducing repetitive tasks can allow the workforce to focus on the operational goals and improve the overall process by reducing downtime, minimizing human errors and saving costs. All these benefits will ensure that adoption of intelligent automation platforms will proliferate along with more intelligent applications across all industries.
Access to the full report requires a subscription to ISG Research. Please contact us for subscription inquiries.