Intelligent Capture Drives AI ROI
PwC estimates artificial intelligence (AI) will contribute up to $15.7 trillion to the global economy by 2030. That enormous value creation comes from productivity gains resulting from automating business processes and augmenting the existing labor force, as well as increased demand for higher quality and more personalized AI-based products. I recommend the PwC report to everyone interested in AI to better understand the impact of it on different industries and the innovative list of hypothetical technological solutions that will shape these industries in the future.
For AI’s impact today, we can look at Forrester’s list of the 10 hottest AI technologies. The leaders in terms of investments are the following: decision management (traditional rule-based business process management), AI-optimized hardware, machine learning platforms (APIs, SDKs for ML training) and virtual agents. However, among this innovation there is still the valid question, “How do you measure ROI from AI investments?”
Bharath Kadaba, CIO of Intuit, put it this way, “You can fall into signing up to solve a problem that's hard to solve, and you can put money behind it and you may not get any clear results.” And the recommendation is: “For all the intense interest in AI investments, the ROI for AI is not well-understood. Rather than make bad guesses, CIOs should treat AI projects like venture capital investments.”
Well, CIOs are not venture capitalists. Their job requires low risk tolerance and the mindset that, “Nobody ever got fired for buying IBM.” They are pragmatic and understand the business value of systems they implement. AI with a clear ROI is the ideal combination—as long as you know which technologies can pave the company’s way into the AI-powered economy, while enabling a strong ROI early in the project.
AI ROI You can Measure
There are early use cases where AI technology has delivered a quantifiable ROI, for example personalization technology for the B2C market. In these cases, AI can learn customers’ preferences from their choices and deliver a customized product or service. Another example is testing or simulation in the manufacturing industry where testing of a final product is extremely costly, for example, in building aircrafts. However, these technologies are niche and specific for a particular industry.
Intelligent capture is another example of AI technology with proven ROI. It converts unstructured information into structured data. For example, it can take content from documents both electronic and paper, or digital communications such as emails and faxes, and make it available in any structured format, including database records, XML or JSON files. By transforming the data into a structured format, they are easier to analyze and knowledge workers can use the data to make faster and smarter decisions within automated business processes.
What is so special about this AI technology is every organization works with documents and unstructured content and can benefit from the intelligent capture. And what is even more important—capture enables data not just for ongoing transactions, but also for using in the machine learning algorithms.
Machine Learning Enables AI
AI requires training, training requires digital data, and capture feeds AI with data. For example, if you want to train AI to control and manage expenses, you need to make sure expenses are properly recorded and all incoming invoices and receipts are processed accurately and timely. This is exactly what capture does—prepare the accurate data. Modern intelligent capture technology extracts data from billions of forms, financial documents, medical records, IDs and other sources in thousands of organizations. Efficiency of this transactional data extraction is proven by strong ROI in terms of early payment discounts, no late fees and more timely payments.
AI Increases the Value of Data
Before the digital transformation uprising, the value of capture technology mostly came from streamlining operations and saving processing costs. It explains why capture was used mostly in mission critical transactions. Now, the value of data has grown dramatically due to unlimited opportunities of using AI for finding new insights and creating a competitive advantage from the same data. Previously, a driver license was used for validating customers’ identities, now companies can leverage information about customers’ height, weight and hair color to personalize their services. And if some information was not captured before but still sits in document archives, systems can now go back and extract additional data.
With data being critical to driving ROI with AI, make sure the capture technology used is flexible enough to support growing requirements from future use cases. It should at least be easy to start processing content from a new source, an additional document type or capture one more field from the same document. By leveraging data more effectively, organizations can better deploy AI applications to compete in future markets.