Home / News and reports / Blog Posts / Understanding the Story behind the Data. An Introduction to Process Mining with Celonis

Understanding the Story behind the Data. An Introduction to Process Mining with Celonis

Posted 30 June 2020 by Ellen Ashmore

In this blog post, C2D3 are excited to introduce our members to Celonis. In June 2020, Celonis presented a fantastic webinar linking the industry benefits of process mining to academic research, through their Academic Alliance team. The webinar is available to watch (access below) and this blog runs through some of the most frequently asked questions in the field of process mining.


Celonis cover slide

What will you learn in the webinar? 

  • Insight into process mining
  • How we can use data science to understand processes
  • Where we can find process in the everyday world and within research
  • Introduction to the process mining tool
  • Theoretical and practical learning
  • Live demonstration of the software


Webinar access 

The webinar is available for our members to watch, your C2D3 login will be required to access the video.



Angela-Sophia Gebert works as Academic Alliance Manager at Celonis, where she helped to build up the global education and research programme with a particular focus on bringing Process Mining technology to the UK academic world and science communication. She graduated in Linguistics, Psychology and Economics from the University of Oxford and LMU Munich. Prior to joining Celonis Angela has worked in the financial services and consulting industry for Savills Investment Management and Bain & Company. Since then she has given over 50 guest talks and workshops on Process Mining.


Unicorn company

The story of Celonis

Celonis was founded in 2011 by three students from the Technical University in Munich, studying mathematics and informatics. From volunteering on a student consultancy project optimising service desk requests, the idea for Celonis was formed; to use existing data within a company and data science to create a digital footprint for an organisation to reconstruct processes.

The idea of process mining stemmed from existing academic research and successfully turned into practice by Celonis, through industry-ready software creation. The company has transformed from a Start-up venture to Unicorn company, with 200 % growth every year, with 800 employees around the world headquartered in Munich and New York. With a valuation now at $2.5 billion, Celonis identifies themselves as a Pegasus company, able to fly on their own, through positive cashflows year after year.


Celonis industry success

Industry success

Process mining is applicable and successful in all industry sections including manufacturing, high tech firms; telecom, media and entertainment; energy and chemicals; consumer, retail and wholesale; pharmaceuticals and life sciences; financial services and insurance; and aerospace and defence and the biggest players in those fields like Vodafone, Telecom, Uber or Unilever have already adopted Process Mining as part of their digitalisation strategy and daily operations.



What is a process? 

A business process or business method is a collection of related, structured activities or tasks that in a specific sequence produces a service or product (serves a particular business goal). Different activities taking place at different points in time, leading to a certain end result. There can be deviations in the process, for example, cancelled orders, wrong product delivery, errors or defaults in manufacturing. A large-scale enterprise can see large-scale cost implications from problems and issues in business processes.

What else is a process? 

There are many examples of processes outside of industry or business activities. Some examples include: 

  • Biology: the growth and development of forests, using process mining to reconstruct the growth of forests through the monitoring of key parameters.
  • Community and social behaviour: decision making, and repetitive systems can be seen as processes. 
  • Healthcare: planning surgical procedures, the flow of people through a healthcare system. 
  • Websites: interaction with a website and the journeys made through the website. 
  • Linguistics: what happens in sentences, different words taking place at different points in time.

How do you get insights into business processes? 

There are many ways to gain insights into business processes, and these can be characterised by how insightful they are. Some of these activities will provide information quickly, but other activities (e.g. interviewing and consulting) are slow and you end up with just a snapshot of the as-is process. 

Old process mapping techniques were subjective and partial data, lengthy and costly, and a snapshot of one time point. This leads to making large presumptions through the simplification of information. 

Process mining in turn is data-based, objective and can deliver complete and immediate insights through real-time data upload. 

What is process mining? 

It is the combination of two disciplines: Process Science and Data Science. 

Process Science is process management, automation, process control, process improvement, and operations management. 

Data Science: data mining, statistics, machine learning, databases, predictive analysis. 

Process-related thinking dates back centuries but this was taken to another level with the digitisation of businesses, evolving into process mining technology, further advanced by academic theory.

Where can you find process data? 

Process data can be found everywhere there is a digital footprint. Some examples include payments (contactless transactions), website logins, online shopping (browsing journeys), mobility (app-based access to publicly accessible e-scooters and bicycles), online customer support with chatbots. 

What is process mining technology? 

Process mining technology takes real-world activities, system support tools (IT systems, client management systems, purchasing systems etc.), the generation of an event logs from the systems, and uses this digital footprint information to conduct process mining. The event log contains a minimum of three pieces of information: a case ID, the activity name, and a timestamp. 

What can process mining do? 

  • Look at data over time
  • Identify correlations 
  • Pinpoint root causes 
  • Make predictions 
  • Open application frameworks – anything you can do with Python, you can apply to process mining 

Can I see a demonstration of the software? 

A live demonstration of the software took place in the webinar. The demonstration starts at 29:51 into the video.

Can I access the academic licence edition? 

To create your own software licence and access the Celonis’ academic software for research: 

Please create an account with your academic e-mail address. You can use the academic licence for any research now or in the future. The academic software must not be used for commercial use.

Can I try the software with practical example? 

We tried some first-step discovery to understand a pizza delivery process and its bottlenecks together with Celonis. Here you can find the case. 

“The Pizzeria Mamma Mia is selling take-away pizza. The business is generally going well. However, recently they realized that their customer satisfaction is comparatively low and that they are making negative profits for some of their deliveries. They want to find out the reason by using process mining technology. Your task is to help the Pizzeria Mamma Mia get deeper insights into their processes and how they are affecting critical KPIs.” 

  • Try the Pizzeria Mamma Mia progress mining challenge.
  • The demo dataset is provided in the academic software licence. 
  • The challenge is described at 46:30 minutes in the webinar. 
  • A group discussion of the challenge findings were discussed at 58:16 minutes in the webinar.

Where can you see process mining in real life? 

  • German airline Lufthansa are using process mining to look at their ground operations, particularly activities which influences departure times (punctuality, what causes delays etc).
  • Supply chain bottlenecks with real-time analysis for the COVID-19 pandemic.

Is R integrated in the software? 

Yes, both R and Python are the two main open applications linked into the software. This requires a good knowledge of your processes; your R or Python script connects into the Celonis software with a plug-in. 

What computing power is available when working with the cloud software?/What amount of data can the software deal with? 

Celonis handles 20+TB data at Siemens from over 70 different source systems. This really is big data. What is good to know is that the data is highly compressed (dumped as compressed files on the source system without impacting the source system’s performance) before it is actually being transferred. 

With the free academic software version you are entitled to upload up to 10GB.

What features are built into the software to deal with personally identifiable data and how is the data anonymised? 

Celonis holds the ISO27001 certification as an information security standard. Furthermore personal information can be e.g. pseudo-randomised or certain columns containing personal information can be removed from the database before engaging with the software. One example of where this is particularly critical is healthcare data containing patient information.

Does Celonis fund PhD studentships? 

This depends on the project. Where there is a common research interest from both parties this is a possibility. 

What else does the Celonis Academic Alliance team do? 

The Celonis Academic Alliance provides software access, case studies and other teaching materials, workshops and teaching support for university-level education in their field. They now have a network of 300+ academic partners in the UK and over 25,000 academic software users! They also have an R&D programme to engage and collaborate for research projects or to partner on research grant applications. 

To find out more and engage with Celonis, you are welcome to contact C2D3 who put you in touch with Angela from Celonis. 

About us

The Cambridge Centre for Data-Driven Discovery (C2D3) brings together researchers and expertise from across the academic departments and industry to drive research into the analysis, understanding and use of data science and AI. C2D3 is an Interdisciplinary Research Centre at the University of Cambridge.

  • Supports and connects the growing data science and AI research community 
  • Builds research capacity in data science and AI to tackle complex issues 
  • Drives new research challenges through collaborative research projects 
  • Promotes and provides opportunities for knowledge transfer 
  • Identifies and provides training courses for students, academics, industry and the third sector 
  • Serves as a gateway for external organisations 

Join us