The business case for hyperautomation

Authentic technology revolutions are easy to spot because they are also economic revolutions.

Hyperautomation, pairing humans with AI and RPA bots to improve the speed and accuracy of work by a factor of 10 or more, is one such revolution. Done right, it unlocks entirely new business models and will drive economic and social change beyond our current imagining.

Hyper-hype

Covid-19 has put the idea of hyperautomation – of complete automation at speed – to the test. We have already seen seismic changes in the patterns, methods and tools of how we work. That change is here to stay; by 2030 we should expect the manual, paper-based business processes we left behind in March 2020 to be as anachronistic as fax machines and Rolodexes.

From the steam engine to the silicon microchip, whenever incremental changes are made to reduce the cost of production by a factor of ten or more, new business models suddenly become possible. Lighting your house with oil extracted from whales killed on the other side of the world only stopped when trains could deliver Pennsylvania oil for less. And as energy prices fell, entire new industries and cities grew across the land – vast economic changes that can help enable social progress as well.

Appian CEO Matt Calkins coined the term “Inverse-Moore’s Law” to describe the effect of cutting in half the time and cost it takes to build applications and automate work, replicated every two years. It comes from Intel CEO Gordon Moore’s famous observation that, as the number of transistors that could be incorporated into a microprocessor doubled every two years, performance should rise and prices fall accordingly. He was right. Low-cost and high-volume microprocessors didn’t just put a computer in every home, they changed how we work, live, and play.

The cost of paper

If you want to radically improve customer experience, you’re going to have to find your own 10x and 100x automation goals. One easy place to start that comes up with the CIOs I speak with is paper. A “paperless office” is now a decades-old promise, which has left most enterprises with an unfortunate legacy of millions of PDFs, spreadsheets, Word docs, and error-prone OCR workflows. The result is that most organizations are still drowning in paper. Even if it is “digital” paper, it’s still a mass of unstructured data that can’t be easily worked on.

Scanning documents with OCR can be expensive – it’s predicted to be a $12.6bn industry by 2025 – and involves plenty of human intervention to set up and troubleshoot errors every time a form changes. That all adds up… and that’s before the “work” has even started.

Take an economic view of the problem. In this case, look at price-per-page, including the software costs, human costs, and time spent working through errors in lost productivity. For instance, if you have to update 200,000 legacy contracts with new LIBOR rates and each contract has upwards of 80 pages, your price per page is the time it takes lawyers and clerks to process and validate roughly 26,000,000 pages using OCR and any workflow technology you have. Let’s say an error takes 10 minutes for a human to find and correct. Each percentage point of error you can reduce adds up to millions of dollars in savings.

A hyperautomation approach to the same problem uses low-code integration and Robotic Process Automation (RPA) to bring documents from different sources into the same workflow. Documents can then be classified and the data extracted by AI, even from complex forms with tables, checkboxes, and handwritten notes. When the AI detects an error, the system automatically offers the suspect data to a human for validation or correction, and the system learns from each of these interactions to improve over time. By maximizing the straight-through processing, pretty soon you are helping ten times the number of clients with greater accuracy and quality than ever possible before.

The case for platforms

Hyperautomation can be achieved in different ways, but there are three significant advantages to using a low-code automation platform: overcoming the scarcity of skilled computer scientists, future-proofing your investment in hyperautomation, and baking in resilience in your systems and processes to deal with future crises.

Using a low-code automation platform to build strategic apps and automations allows you to increase productivity from your existing skilled developers, while also opening up possibilities for hiring non-traditional developers. It also insulates you from skills shortages and provides AI-assisted DevSecOps to protect you from your least experienced or most malicious developer – inside or outside your organization.

Low-code automation also reduces the amount of technical debt your team is taking on with each app build. By eliminating most of the cost that would otherwise be required and rendering a designer’s intent directly in the platform, you are future-proofing your investment in hyperautomation.

Each update to the platform should also be able to upgrade your apps and automations without costly delays to rewrite the code. For instance, if the platform upgrades with a new Javascript framework, every app built with it should get the same boost to user experience. iOS and Android upgrade each year and basic deployment infrastructure is being revolutionized with containerization and AI, narrowing the viable business cases for open source and custom code vs. platform development.

Hyperautomation is now

In 2020, the term “hyperautomation” topped Gartner’s list of strategic technology trends for the first time, but we’ve been working on laying the foundation for this 10x economic revolution since we founded Appian in 1999.

Empower your people with low-code so they can use AI and RPA today and your own 10x workflow improvement through hyperautomation is within your reach.

Michael Beckley, Founder and Chief Technology Officer, Appian

Source link