Intelligent Automation of Manual, Document-Based Workflows Creates Huge Value Opportunities for Enterprises
The hype around artificial intelligence (AI) is exciting but it also creates a lot of confusion and skepticism in the minds of both data scientists and their business colleagues. Despite its promise, and its growing adoption, there is still too little users can point to in terms of real business results. This is especially true when it comes to the plethora of unstructured and document-based content that makes up more than 80% of the data in most enterprises.
Working with clients in banking, insurance, industrial services and manufacturing, Indico, a provider of enterprise AI solutions for intelligent process automation, has identified some of the key considerations and characteristics that separate successful AI initiatives from unsuccessful ones. It has also outlined a framework to help enterprise users evaluate opportunities within their own organization and increase the likelihood of tangible and measurable ROI.
“If we put aside all the hype around AI, we’ve seen tremendous progress in the ability to deploy the technology to automate manual, document-based work processes and drive valuable business results,” said Tom Wilde, CEO of Indico. “If we understand the real capabilities of these tools, and work with the business to identify the right use cases and workflows to apply them against, we can generate significant ROI in a tangible and practical way.”
Three critical success factors for any AI initiative
Indico has identified three core factors that every initiative must address to be successful:
- Data Access & Training Data – Users need a well-formed set of inputs and outputs of sufficient volume to make even the most basic machine learning problems tractable. And the training data must provide quality examples of the desired outcome. Data requirements vary based on the complexity of the problem, but a standard dataset size for training a model can be between 10,000 and 100,000 labeled examples. This is typically where many implementations fail. Fortunately, newer approaches to AI use frameworks such as Transfer Learning to dramatically reduce the amount of training data required – in some cases, to just a few dozen examples.
- Data Science and Line of Business Collaboration – As is the case in most technology initiatives, business users and technical staff need to collaborate effectively to produce the desired outcomes. This is especially true with AI where the subject matter experts (SMEs) play such a critical role in the definition of success, and the underlying technology is so complex. With AI, the importance of data science expertise is typically well understood. However, the role of the business SME is often undervalued.
- Identification of High-Potential Use Cases – The most vital component for realizing ROI is a clear understanding and definition of a desired outcome. This enables the project team to work backwards in terms of identifying the steps that can be augmented, enhanced or automated, the data available, and a set of previously identifiable outcomes that can be used for training the models. Is the goal to fully automate a process, or augment a manual process? These carry different considerations and ROI implications.