As enterprises adopted RPA solutions to automate specific tasks and processes, requirements arose for other capabilities, such as intelligent document processing (IDP), API-led integration, and task mining, to automate end-to-end processes. This evolution simply indicates an increased capability of these solutions (along with adjacent products) to support a greater degree of automation for an entire end-to-end process.

Bot stores offering prebuilt components and RPA bots for automating a specific task are now a key requirement for enterprises that have scaled beyond the first couple of dozen RPA bots. API-led integration tools are used when application and data integration can be achieved via APIs. Beyond legacy optical character recognition (OCR) tools, RPA solutions integrate with IDP solutions to enable document ingestion and processing. In other words, they convert semi-structured and unstructured data in documents to a structured format. Task mining empowers organizations to understand how tasks are performed by monitoring user interactions with their workstations. Autogeneration of process design documents (PDDs) for automation opportunities identified via task mining helps RPA developers expand their understanding of end-to-end processes.

While RPA solutions have commoditized to some extent, the overall intelligent automation platform (IAP) market is growing fast, with several new technical developments. RPA vendors have quickly added new tools and capabilities (mostly via acquisitions and partnerships) to address the requirements of intelligent automation use cases. With RPA scalability and bot resiliency being a recurring question, a few RPA vendors have put in efforts to address these issues. We see that lines of demarcation between integration and process automation approaches are blurring, with the traditional friction between deep integration—API-led integration and dedicated application connectors provided by integration platform as a service (iPaaS)—and surface-level automation lessening over the last few years.

There is a greater propensity for RPA solutions to call dedicated AI/ML APIs (as in computer vision APIs) to meet specific requirements of a use case. Another interesting observation is a greater number of RPA implementations have scaled to hundreds and thousands of RPA bots, indicating that RPA, when applied to the correct use cases and governed properly, can deliver positive outcomes. Issues arise when RPA initiative leaders spend too little time on analysis and optimization; they often think of RPA as a hammer and then see most of their business processes as nails to hit. Needless to say, automation of an inefficient business process with RPA will deliver only minimal gains.

This GigaOm Radar report highlights key RPA solution vendors and equips IT decision-makers with the information needed to select the best fit for their business and use case requirements. In the corresponding GigaOm report “Key Criteria for Evaluating RPA Solutions,” we describe in more detail the key features and metrics that are used to evaluate vendors in this market.

The post GigaOm Radar for Robotic Process Automation appeared first on Gigaom.

While there are other tools available to generate REST APIs against legacy back ends, RPA solutions are preferred because of their less-technical approach to automation and faster time to automation. RPA empowers technology savvy business users who are not part of a technology team—finance, HR, operations, or any other department in the organization besides IT—to create and own automations. This degree of accessibility places control of the solution and outcomes into the hands of the people performing the manual tasks.

As enterprises adopted RPA solutions to automate specific tasks and processes, requirements arose for other capabilities—such as intelligent document processing (IDP), API-led integration, and task mining—to automate end-to-end processes. This evolution indicates an increased capability of these solutions (along with adjacent products) to support a greater degree of automation for an end-to-end process.

Just as we’ve seen in other product lines, marketplaces are now part of RPA solutions. These RPA specific marketplaces are called “bot stores” and offer prebuilt components and RPA bots for automating a specific task. These bot stores are now a key requirement for enterprises that have scaled beyond their first couple of dozen RPA bots. API-led integration tools are used when application and data integration can be achieved via APIs.

Beyond legacy optical character recognition (OCR) tools, RPA solutions integrate with IDP solutions to enable document ingestion and processing. In other words, they convert semi-structured and unstructured data in documents to a structured format. Task mining empowers organizations to understand how tasks are performed by monitoring user interactions with their workstations. Auto-generation of process design documents (PDDs) for automation opportunities identified via task mining helps RPA developers expand their understanding of end-to-end processes.

RPA solutions have gradually expanded to offer a greater set of features and capabilities than mere task automation could. RPA adoption has entered the mainstream, and some RPA implementations have scaled to thousands of bots. At the same time, some RPA initiatives fail to scale beyond the first couple dozen bots. Such failures cannot be simply attributed to the technical aspects; people and process aspects are equally important for RPA initiatives to scale and deliver positive outcomes.

This GigaOm Key Criteria report details the criteria and evaluation metrics for selecting an effective RPA solution. The companion GigaOm Radar report identifies vendors and products that excel in those criteria and metrics. Together, these reports provide an overview of the category and its underlying technology, identify leading RPA solutions, and help decision-makers evaluate these solutions so they can make a more informed investment decision.

The post Key Criteria for Evaluating Robotic Process Automation Solutions appeared first on Gigaom.

Traditional integration approaches—including enterprise service bus (ESB)/service-oriented architecture (SOA), enterprise application integration (EAI), and custom-code development—proved to be ill-suited for meeting modern hybrid integration scenarios in an agile manner. Given the increasing need for organizations to do more with less, integration centers of excellence (COEs) have embraced agile approaches to integration, and it is no longer uncommon to see line-of-business (LoB)-led iPaaS adoption.

In certain cases, iPaaS is used as a means to deliver packaged integration for a vendor’s own SaaS applications or those provided by a specific set of independent software vendors (ISVs). For the purposes of this report, we are not considering those iPaaS solutions in which integration capabilities are limited to cloud services brokerage (CSB)/integration brokerage arrangements.

The GigaOm Key Criteria and Radar reports provide an overview of the iPaaS solutions market, identify capabilities (table stakes, key criteria, and emerging technology) and evaluation metrics (non-functional purchase requirements) for selecting an iPaaS solution, and detail vendors and products that excel. These reports give prospective buyers an overview of the top vendors in this sector and help decision-makers evaluate solutions and decide where to invest.

The post Key Criteria for Evaluating Integration Platform as a Service Solutions appeared first on Gigaom.

iPaaS solutions provide a centralized console for scheduling, monitoring, and managing integrations. Only multitenant, scalable, cloud-based integration solutions that provide the necessary tools and dedicated resources for faster development of integration flows—and offer the requisite data security and governance for such integrations—are iPaaS solutions. Developer productivity tools, such as a drag-and-drop approach to integration-flow development and prebuilt connectors and templates, are foundational components of the iPaaS value proposition. In terms of data security, key capabilities include transport layer and application- and network-level security and support for implementation and administration of governance policies.

Over the past few years, several iPaaS vendors have extended the capabilities of their solutions to include support for application programming interface (API) management, master data management (MDM), B2B/electronic data interchange (EDI) integration, and mobile app/back-end integration (hybrid integration use cases). This is not surprising given the flexibility to consume these solutions as a cloud service (PaaS) and their ease of use, which provide greater business agility and a proposition to lower total cost of ownership (TCO).

As organizations are increasingly tasked to do more with less, integration centers of excellence (COEs) have embraced agile approaches to integration, and it is not uncommon to see line of business (LoB)-led iPaaS adoption.

iPaaS is firmly established as a significant approach to cloud integration, and there are two reasons for its growth. First, iPaaS adoption in many enterprises is driven by LoB, and once IT is conversant with the features and functionality of the solution, the use of iPaaS is extended to other integration use cases. Second, several iPaaS vendors have expanded the features and capabilities of their solutions to cater to the needs of MDM, B2B/EDI integration, and API management.

It is increasingly important for iPaaS vendors to target new user personas and a broader set of integration use cases (such as B2B/EDI integration, API lifecycle management, data integration and data management, and mobile application and back-end integration). In this context, there are two noteworthy developments: self-service integration capabilities that enable non-technical users (citizen integrators), and machine learning (ML) capabilities that simplify development of integration flows and connect various applications and/or data sources.

iPaaS vendors drive significant innovation targeting mid-market organizations (with revenue less than $1 billion) who focus on developing low-code integration capabilities even for B2B/EDI and event-driven integration use cases. Integration marketplaces have rapidly evolved to offer hundreds, if not thousands, of prebuilt recipes and templates and, increasingly, components for automation of standard processes (for example, order-to-cash and procure-to-pay processes). Such vendors are democratizing integration, and the lines are blurring between application integration, data integration, and workflow and process automation. Mid-market organizations can leverage a mix of hybrid integration capabilities at a lower cost of ownership using such iPaaS solutions.

This GigaOm Radar report highlights key iPaaS vendors and equips IT decision-makers with the information needed to select the best fit for their business and use case requirements. In the corresponding GigaOm report “Key Criteria for Evaluating iPaaS Solutions,” we describe in more detail the key features and metrics that are used to evaluate vendors in this market.

The post GigaOm Radar for Integration Platform as a Service appeared first on Gigaom.

A low-code process automation platform offers visual development tools that enable both skilled and citizen developers to analyze, model, optimize, and automate business processes. In particular, its drag-and-drop interface enables citizen developers to incorporate changes in processes and deploy them rapidly without needing to write complex code. When consumed as a cloud-native platform as a service (PaaS), users can experience several benefits, including better scalability and auto-scaling, high availability (HA), improved security and compliance, and continuous delivery and instant deployment.

The post Low-Code Process Automation appeared first on Gigaom.

Over the last decade or so, several vendors experimented with a combination of Python code and optical character recognition (OCR) in what was termed “intelligent OCR.” While there was an increase in data extraction accuracy (+5 to 10%), the results were still not in line with the requirements of IT, automation centers of excellence (CoE), and process leaders.

IDP solutions involve pre-processing and post-processing stages: the first primes the document in terms of shape and size, and the second ensures that a greater degree of accuracy in processing unstructured data can be achieved.

The availability of AI/ML/NLP capabilities in IDP ensures that the strike rate in processing data is significantly higher than a mere OCR tool can deliver. IDP tools, in principle, are supposed to function with minimal training in terms of minor template changes, whereas with OCR, this type of flexibility doesn’t exist.

Pre-built AI/ML capabilities and business rules enable automated verification and validation of data, as well as continuous learning and improvements based on AI/ML algorithms and user input. IDP combines OCR, data capture, and AI/ML to automate the retrieval, understanding, and integration of documents required for executing a business process. Know your customer (KYC), invoice processing, insurance claims, patient onboarding, patient records, proof of delivery, bills of lading, and order forms are some of the key use case document types for IDP solutions. IDP software can be useful in industry-specific processes, such as customer onboarding, mortgage processing, trade finance, and legal documents. Some of the key buyers of IDP solutions are leaders in a number of business areas: shared services, process, automation CoE, global business services (GBS), digital transformation, and finance.

Decision-makers will want to determine whether an IDP solution is a refurbished intelligent OCR tool or a dedicated product built from the ground up to meet the requirements of complex use cases. IDP solutions offering a template-free approach can be trained on new document types and layouts by using a certain number of sample documents; such solutions will usually deliver a greater value at a lower cost. IDP solutions should be adopted as part of a broader enterprise automation strategy in which suitable software products are used for meeting various task, process, and document automation requirements.

There are more than 60 vendors that provide IDP-type software capabilities. In this GigaOm Radar report, we focus only on vendors with a noteworthy footprint in terms of scaled IDP implementations where the IDP product involves more than an add-on to a broader robotic process automation (RPA) or intelligent automation platform. We therefore excluded vendors whose solutions were more service based, as well as those who offer basic OCR-based document capture without significant advancement in AI/ML capabilities.

The post GigaOm Radar for Intelligent Document Processing appeared first on Gigaom.

The emergence of IDP solutions beyond intelligent optical character recognition (OCR) tools resulted from significant investments in AI, ML, NLP, and deep learning capabilities meant to overcome the limitations of legacy OCR tools. Even with relatively bad quality documents and scanners, IDP solutions can deliver greater accuracy in data extraction, something legacy OCR tools frequently struggle to supply.

Decision makers will want to determine whether an IDP solution is a refurbished intelligent OCR tool or a dedicated product built from the ground up to meet the requirements of complex use cases. IDP solutions offering a template-free approach can be trained on new document types and layouts by using a relatively limited number of sample documents, and such solutions will usually deliver a greater value at a lower cost of ownership. IDP solutions should be adopted as part of a broader enterprise automation strategy that uses suitable software products to meet various task, process, and document automation requirements.

This GigaOm Key Criteria report details the criteria and evaluation metrics for selecting an effective IDP solution. The companion GigaOm Radar report identifies vendors and products that excel in those criteria and metrics. Together, these reports provide an overview of the category and its underlying technology, identify leading IDP offerings, and help decision-makers evaluate these solutions so they can make a more informed investment decision.

The post Key Criteria for Evaluating Intelligent Document Processing Solutions appeared first on Gigaom.

The document structures of these invoices were identical to transparently identify how each IDP solution would treat an unfamiliar invoice layout. Both IDP products processed the same document sets. This provision ensures a neutral collection of test documents that neither IDP product is pre-trained to recognize.

We found that the Automation Hero Hero Platform delivered 68% greater accuracy than ABBYY FlexiCapture in terms of global average (“headers” and “line items” combined) for invoice processing. In terms of headers (e.g., invoice number, invoice date, amount, customer name, customer address, etc.), the Hero Platform delivered 67% greater accuracy than ABBYY FlexiCapture IDP product. And for line items in invoices, the Hero Platform delivered 69% greater accuracy than ABBYY FlexiCapture.

For the other use case, handwriting recognition for snippets from actual medical forms, Automation Hero context-aware optical character recognition (OCR) delivered 281% greater full-field accuracy when compared to ABBYY FlexiCapture.

The post GigaOm Performance Benchmark for Intelligent Document Processing (IDP) Solutions appeared first on Gigaom.

The management of the integration infrastructure layer is both specialized and multifaceted and includes monitoring, observability, configuration and security management, alerting and automation, and performance optimization. Extracting intelligence from the integration infrastructure layer is key to achieving the overarching goal: situational awareness (and the capability to act on these insights to achieve IT and business objectives).

Transaction observability includes the ability to determine how software’s internal state changes in response to external outputs, specifically regarding transactions. Using the right combination of monitoring, logging, documentation, and visualization tools, it is possible to develop an appropriate situational awareness of distributed systems and applications. This Sonar report assesses software products that deliver toward the goal of situational awareness, including transaction observability for distributed systems and applications as well as all the elements of integration infrastructure management.

The post GigaOm Sonar: Integration Infrastructure Management and Transaction Observability appeared first on Gigaom.

Figure 1. Process Mining Procedure

Organizations can derive good value by applying process mining to already digitized processes and for which unstructured work (such as reviews and manual approvals) is undertaken outside the IT systems. Using process mining, organizations can analyze their processes from the bottom up. Process mining uses the historical data in IT systems, so there’s no need to have a preexisting process model for analysis.

In simple terms, process mining algorithms determine how a process model is inferred based on raw event data. An algorithm is used to correlate event log data to identify trends and patterns, and aggregate metrics obtained from event logs recorded by IT systems. The models obtained from process mining can then be compared with existing models to check for conformity and discover relatively more efficient process models.

Task mining enables organizations to understand how tasks are performed by monitoring user interactions with workstations. Task mining technologies include optical character recognition (OCR), data mining, pattern recognition, and natural language processing (NLP). Task mining allows organizations to track KPIs of individual tasks that are often manual. Organizations can track task productivity and make data-driven process improvement decisions by analyzing user interaction data. Task mining focuses on analyzing employee interactions with one or multiple applications or information systems (like Microsoft Excel and Microsoft Word) while maintaining data security and privacy for sensitive information. Task mining helps the automation CoE and business leaders by enabling efficient discovery of high-potential opportunities for automation.

Process mining has a broader purview and aims to discover, analyze, and monitor end-to-end processes and sub-processes using event logs from information systems. Process mining is a sequence of events constituting individual steps. Process mining algorithms exploit data from event logs, which can be different types of databases and files from enterprise information systems and enterprise applications—for example, an enterprise resource planning (ERP) application. The data is visualized as a process flow that presents a given process as it’s actually executed.

Process mining solutions can be used as part of digital transformation initiatives to discover or create a transformation pipeline, and monitor progress to achieve key business objectives, such as faster time to market, improved return on investment (ROI), and a better customer experience (CX). With process mining, enterprises can extract hidden process insights to uncover new business opportunities (that is, cross-sell or upsell) and better ways of doing business. Table 1 provides a comparison of task mining and process mining.

Table 1: Task Mining versus Process Mining

This GigaOm Radar report focuses on process mining and task mining solutions. Process and task mining solutions that are tied to a specific automation ecosystem or part of a broader automation suite ultimately aimed at driving robotic process automation (RPA) adoption are not included in this evaluation. In the corresponding GigaOm report “Key Criteria for Evaluating Process Mining and Task Mining Solutions,” we describe in more detail the key features and metrics that are used to evaluate vendors in this market.

The post GigaOm Radar for Process and Task Mining Solutions appeared first on Gigaom.