Business output is directly proportionate to how well teams are supported to do a good job. If they can’t complete a task because they don’t have all the information at hand, teams become steadily less productive. Yet, ironically in this digital age, this situation still prevails across all kinds of enterprises. This is usually because of still-present silos across extended business processes.
Where once the issue was incompatible single-use legacy systems, today it is best-of-breed business ecosystems such as SAP, Salesforce.com, Workday, ServiceNow, Microsoft 365 and more, which, when used together in the same business, create walls between related content and insights.
As a result, despite extensive investment in the latest technology, knowledge workers continue to manually consolidate information from disparate systems to do their jobs. More than 60% of the time, this involves looking across four or more systems, according to SER Group data; in large companies, it could be 10 or more. Where process automation exists, it is typically only partial.
This is not the digital dream that anyone wanted. It does not drive up productivity nor does it delight customers. It does nothing to reduce risk or improve process efficiency. Nor does it help in identifying new business opportunities. It leaves businesses suffering a very poor return on their “new tech” spend.
The limited-knowledge trap: dark data
On top of interdepartmental dead ends, 80% of the data that exists across an enterprise remains unstructured, limiting its onward potential as a source of actionable knowledge. Confined to PDFs and emails, these information nuggets are difficult to call into play as part of wider company “intelligence.”
Meanwhile, more than 50% of information in an organization can be categorized as “dark data”: there is a sense that it exists somewhere but its whereabouts are not easily identifiable and the content is largely unsearchable. If the failure to surface this information results in poor customer service, missed sales opportunities or business accounts being prioritized despite repeated late payments, then the company is likely leaking value.
The answer must be to bring unstructured data into the light and link any embedded insights with existing structured data to complete the narrative about an account or situation, drive more informed decision-making, and facilitate more sophisticated process automation.
Meaningful automation: it’s all about the context
As demand for many goods and services soars to pre-pandemic levels, and organizations strive to do more with the resources they have, they need to be able to manage content in ways that are smarter, more connected and context-aware. And it is here that next-generation, AI-enabled content automation promises to help.
AI technology is advancing all the time, as is general understanding about where automation can add real value for a business. Next generation automation systems will actually understand content, so that they are able to take the appropriate actions in a complex enterprise environment with little or no human intervention.
This is very timely, given the ongoing disruption to the workforce. Now more than ever, smart application of integrated content and process automation has an important role to play in relieving the pressure on overstretched teams.
Appreciating and adding to the enterprise narrative
Next-level automation will see more human-like understanding and contextual memory of what that information is, what it means, how it adds to the story, and how this knowledge could be applied in smarter ways across different use cases, expedite next courses of action, and deliver a broader range of business benefits.
Take the example of invoice and financial management, and the potential for AI-enabled tools to “read” and make sense of incoming documents and intelligently trigger next actions, in the context of wider enterprise services like ERP, CRM, contract management, and so on.
This requires the application of more than one type of AI technology, and it is in blending the different capabilities that we start to breathe life into next-generation content management. To ingest and process invoices automatically, for instance, pattern-matching AI such as deep learning is needed to identify the type of document and its constituent parts, while contextual AI recalls how documents of that class are usually handled – rather like a human’s contextual memory.
But how is all of this possible when departmental silos persist? Today an organization might use the SAP application ecosystem for ERP; Salesforce for CRM, Workday for HR, ServiceNow for customer service, and the Microsoft suite for everything in between. There may be specialist systems in use too, such as Cerner and Epic health record systems in a hospital context.
While this best-of-breed software approach can serve a particular function well, inter-system integration at the content understanding level has become increasingly important as organizations have realized the opportunities they might be missing through a lack of cross-functional insights. (They could be missing out on revenue opportunities, for instance, because the sales team has only been able to see information held within Salesforce and not that same customer’s request in ServiceNow for advice and support.)
Content-rich integration, at a process level, would alleviate this problem.
The rise of the ‘platform’
As companies have sought to be smarter and more holistic in their information management and more ambitious in their business process automation, the merits of a single, unifying platform over multiple discrete applications have risen to the fore.
Deploying an ECM system and related tooling might seem to be the obvious answer, though too often this approach fails to produce a sufficiently timely return on investment. Given the accelerating pace of change, and rate of new disruption to business-as-usual, few organizations can afford to make such a generic and wholesale change before seeing the benefits at a defined process level and in tangible business outcomes.
An emerging option with growing appeal is to have solution suites that can “snap together” and leverage contextual AI for a growing repository of shared wisdom, enabling faster workflow and business outcome delivery.
Take the example of a healthcare application suite for onboarding patients and storing their 360-degree view information, that can be readily linked to a finance purchase-to-pay suite for analyzing invoices and order confirmations, which in turn also connects into the hospital SAP system for patient billing. Similarly, a finance suite might be able to link and share insights with sales and legal, and vice versa. Each specialist solution can do a task well and share its knowledge, enriching the wider view of a customer, supplier, patient or business opportunity.
It’s this vision – of delivering more tangible impact and timely ROI at a functional level, while contributing to a shared higher purpose – that is driving the convergence of a number of adjacent technology fields. These include:
• Enterprise content management, for managing content;
• Robotic process automation and business process management, for orchestrating processes;
• Intelligent document processing, for understanding incoming content; and
• Enterprise content integration, for bridging the content silos and content automation applications.
The converged proposition adds up to intelligent content automation, bringing together AI intelligence with content and process automation.
Orchestrating intelligent content automation
To address next-generation business process transformation requirements, there needs to be an intelligent content bridging mechanism that supports knowledge sharing between different software ecosystems – enabling workflows to be orchestrated across software ecosystem boundaries.
A layer of contextual AI can then promote institutional wisdom, through access to every piece of content and the ability to remember how it was used previously, building on these learned patterns over time as more content is processed.
The key to what happens to all of these insights is the way that different AI capabilities can be called into play. This requires a “composable AI” architecture that can harness and build on any or all of the latest AI frameworks. This is important, ensuring that companies don’t find themselves locked into using particular AI functionality, which later loses its value as it is superseded by something better.
Delivering enterprise-scale process transformation
As enterprises become more ambitious in their process automation and content management plans, and look to drive next-level ROI from their latest tech investments, the focus must shift to cross-enterprise knowledge integration and contextualization.
A wide range of developments — in the global economy, in the reinvention of the workplace, and in the world of technology — are creating the perfect storm for AI’s extended and more deeply integrated role in an organization, in its knowledge management and in its processes – augmenting and accelerating the vital everyday work that human teams are working so hard to stay on top of.
It’s against that backdrop that composable, embedded, contextual AI use will drive the next, more ambitious waves of content-centric process automation — and the potential is considerable.
Dr John Bates is the CEO of SER, which specializes in intelligent content automation (ICA) - the convergence of content management (capturing, storing, searching, archiving, and managing enterprise content), business process automation and AI-powered content understanding.