Could Process Mining Be Bigger Than RPA (Robotic Process Automation)?
science formula and math equation abstract background. Concept of machine learning and artificial … [+]
Eight years ago three college friends in Germany – Bastian Nominacher, Alexander Rinke and Martin Klenk – founded Celonis with 12,500 euros. They saw an opportunity to use process mining to transform the way large companies do business. Celonis was profitable from the start and grew rapidly.
It wasn’t until June 2016 that the company raised its first round of institutional capital ($ 27.5 million from Accel and 83North). Then two years later there was a Series B for $ 50 million.
And yes, this week came the mega-round: $ 290 million. The main investor was Arena Holdings, along with Ryan Smith (Co-Founder and CEO of Qualtrics) and Tooey Courtemanche (Founder and CEO of Procore).
Then what is process mining? And why is it so hot right now?
First, process degradation has its roots in the academic world in Europe, which was developed by Dutch professor Wil van der Aalst. He is one of the world’s most cited computer scientists (with more than 92,000 according to Google Scholar).
He recognized that the analysis of event lots from IT systems can provide valuable insights, for example with regard to the mapping of the actual processes of an organization and their context. This makes it easier and more objective to find problems and bottlenecks and to uncover ways to improve the systems.
Now it was Celonis who commercialized the first process mining. It also helped the company leverage existing open source projects and focus on a limited number of use cases such as finance, supply chain, customer service, and sales / marketing. This strategy was important because effective process mining requires a deep understanding of a domain. For example, Celonis has developed pre-built modules to process certain business outcomes (e.g. for on-time delivery or looking for ways to get paid faster).
“Process mining is a simple idea,” said Rinke. “But it’s difficult to get it right for organizations. You have to collect large amounts of data from all types of IT systems. You also need to go beyond integrations and understand the databases below. And all of them are bespoke. “
A key driver for Celonis was undoubtedly the rapid growth of RPA (Robotic Process Automation). “Often in RPA, your first chances are when you ask people which routines are taking the most time,” said Antti Karjalainen, CEO of Robocorp. “As companies advance their automation journey, data-driven technologies are becoming an important part of identifying opportunities. People may not even realize how their own work relates to work in other areas of the organization, and process discovery technologies can uncover these hidden workflows. “
The Celonis software is not only used for preliminary analysis. This is useful for ongoing monitoring to ensure that an RPA implementation is on the right track.
Go to the next level
With the capital injection, Celonis plans to expand sales and marketing. But there will also be relocations to other areas.
“I think our business can be bigger than RPA,” said Rinke. “RPA is mostly about automating tasks and is difficult to scale. However, our approach is more holistic. “
Process mining can also be a gateway for digital transformation, for example with AI (artificial intelligence) and ML (machine learning). “By analyzing large amounts of data, we can filter out the noise and make better predictions,” said Rinke. “For example, when an order comes in, we can tell with 85% to 95% accuracy whether it’s on time or not.”
Regardless of whether the process degradation exceeds the RPA or not, the growth seems to continue for some time to come. Gartner estimates between 3X and 4X for the next two years.
“The Celonis funding demonstrates once again that Process Intelligence is a must for companies undergoing digital transformation and adopting RPA,” said Scott Opitz, President of ABBYY Process Intelligence. “We see this as a positive dynamic in the marketplace as it alerts the C-Suite that they need to know which processes need to be automated first and why, where RPA bots need to run, how they work, and which are quantifiable Benefits they expect to provide these digital employees in the company. “
Tom (@ttaulli) is the author of the book “Fundamentals of Artificial Intelligence: A Non-Technical Introduction”.