What Are Process Intelligence Tools And How Do They Work?

Process Intelligence combines everything from business process analysis, to process improvement and monitoring, based on real operational data. But what exactly are the tools and technologies that comprise Process Intelligence? And how do they work? Let's get into it.

What are Process Intelligence tools?

Process Intelligence tools use data to create transparency into how business processes run, as well as technologies to unlock value and increase performance. When you're looking at vendors selling Process Intelligence software, you're usually looking at a packaged set of tools, technologies, and methods such as:

Process mining: You can think of process mining like an x-ray of your business. It uses business data to visualize, analyze, and optimize your business processes, giving you end-to-end transparency about the real way your organization operates.

Task mining: Similar to process mining, task mining uses data to understand how tasks are executed. But instead of business data, it looks at user interaction data including all the steps that happen outside of major systems, like checking emails or consulting spreadsheets.

Process modeling: The models created via process modeling are used to design an ideal process, taking into account institutional knowledge about how your business operates (such as enterprise architecture, organizational structure, or process landscape). Models can cover a wide range of processes from business processes like Purchase-to-Pay, to customer journeys or even an entire org chart.

Digital Twin of your organization: A living, system-agnostic replica of your operational reality. When powered by the Celonis Context Model, this digital twin combines hindsight, insight, and foresight (via Decision Intelligence) so you can understand operations in depth, simulate changes, and predict outcomes.

Generative A.I and machine learning: Within Process Intelligence, Generative A.I acts as an interface, while Decision Intelligence provides the foresight. By anchoring A.I agents in the Celonis Context Model, non-technical employees can query processes using natural language, receive predictable, cost-effective recommendations, and trust A.I agents to execute actions safely.

Automation capabilities: Whether we're talking robotic process automation (R.P.A), automated workflows, or alerts, a good Process Intelligence tool will come with automation capabilities that allow you to interact with your systems and technologies across your wider organization, such as your E.R.P or C.R.M systems. This allows you, for example, to automatically and directly unblock orders, update master data, or notify employees all from within the Process Intelligence tool.

What value does a Process Intelligence tool bring to an organization?

Processes are at the heart of how things run in your company. By improving your processes you can increase efficiency, better adapt to changing demands, and be strategic about implementing technology.

But most companies struggle to understand how their processes really work. Their systems don't play well together, and their departments don't speak the same language.

In real life, this can look something like this:

A company who wants to improve their Lead-to-Cash process needs to investigate their sub-processes of Order Management, Inventory Management, and Accounts Receivable.

Order Management works with Systems like sap or Salesforce and is concerned with, well, orders. But depending on what system they use; orders aren't always called “orders”. Where Salesforce uses “order”, sap uses V.B.A.K. And that's just one department — the same is true for Inventory Management and Accounts Receivable.

So while all three departments might share the goal of optimizing Lead-to-Cash, they can't collaborate to achieve that goal as they work in siloed systems, with their own terminologies and conflicting concerns regarding performance.

So, clearly, finding common ground to drive performance and value is hard. That's where Process Intelligence tools—anchored by a unified context layer like the Celonis Context Model—make all the difference by providing a single source of truth across siloed systems.

How do Process Intelligence tools work?

Process Intelligence tools do three things:

They bring siloed data together to deliver end-to-end transparency into the processes and workflows that run in an organization, like the steps and parties involved, dependencies between processes, as well as up and downstream effects.

They provide the visual aide to share insights and create a commonprocess language to break down the barriers between systems, processes and departments.

They provide the tools and knowledge to optimize performance and unlock value opportunities across an organization.

They combine Process Data and Business Knowledge to show what happened and why.

They leverage Decision Intelligence (predictions and simulations via L.G.M's) to anticipate future risks and opportunities.

They safely orchestrate human and A.I agent actions across systems.

And they do these things in a cycle of continuous improvement.

It all starts with bringing data together as easily and efficiently as possible. We're talking pre-defined analyses based on process knowledge, and data that works system agnostic,which means from any data source whether it's an E.R.P system, Cloud System, Data Warehouse, or custom-built app.

This data is the basis for the actual process analysis, meaning the visualization and presentation of the data in dashboards, and the analysis of root causes behind untapped value using the Celonis Context Model, which equips A.I models with the exact operational ground truth needed to reason correctly without hallucinating.

The next step is process improvement through automation, streamlining, and simplification. This includes setting up alerts when important K.P.I's fall out of range, but also building and implementing automated workflows for tasks that don't need human intervention. A good Process Intelligence tool will allow you to take action across your systems and platforms and even third-party apps, but without disrupting your business.

Once you've implemented optimization measures, you can use process monitoring to understand the impact of your measures, adherence to process models, conformance, and the development of K.P.I's.

One final thing to note is that Process Intelligence is a cyclical initiative that allows you to continuously drive optimization and unlock value. And it doesn't just apply to one department or process – you can use it to scale process optimization across your organization.

The Celonis Context Model

Enterprise A.I has blind spots when it comes to how your business runs.

The Celonis Context Model creates a dynamic, living, digital twin that mirrors the reality of your supply chain. Combining hindsight, insight, and foresight, it gives your people and your A.I agents the operational clarity they need to reason correctly, decide sensibly, and act reliably.

The Celonis Context Model provides operational context through a dynamic, real-time digital twin of operations, translating the reality of the business into a language that A.I understands.

Combining process data, business knowledge, and intelligence, the Context Model gives your people and Enterprise A.I the operational clarity to reason correctly, decide sensibly, and act reliably.

Understand your operations

Complex operations in Supply Chain and Finance run across dozens of disparate systems, applications and devices. The Context Model integrates data from all of these into an agnostic digital twin of your operations.

The Context Model understands the relationships between all of the documents, materials, and people that make up your business and how they're interconnected and interdependent. It encompasses both the current state of your operations and the full backstory of every step, interaction and decision that led to this moment.

Enriched with business knowledge and intelligence

The Context Model is enriched with business knowledge, the institutional know-how essential to every company, defining goals and objectives, how you work with customers and partners, industry best practices, and crucially, constraints and guardrails so A.I stays on mission and in bounds.

With this foundation, it offers intelligence. Process intelligence, which tells you how your business runs and how to improve it. And Decision Intelligence, which provides predictions about what needs to happen next, and simulations of each scenario to make sure you achieve your goals. With this intelligence, your agents can both fix problems and prevent them altogether.

Open, extensible and future proof

The Context Model is designed as an open and extensible layer that you and your partners can continuously enrich with additional data, business knowledge, and intelligence functions.

Its open architecture allows organizations to integrate any data source, A.I model, or agent while avoiding vendor lock-in and preserving their operational context as technologies evolve.

Process Mapping versus Process Mining: What's the Difference?

In 1921, the first business case for process mapping was presented to the American Society of Mechanical Engineers. Engineer Frank Bunker Gilbreth, and his wife Lillian, argued that having a visual “record of present conditions” would provide businesses with a valuable indicator of “profitable changes” across Production, Sales, Accounting and Finance. You can hear their original arguments in archive footage of one of their presentations, “The Quest for the One Best Way.” It's a fascinating watch, and a rare window into the history of business process management.

Except for one thing… strictly speaking, it's not really history.

That's because today — over 100 years later — around 65% of businesses are still at the first stage of process maturity, where process mapping is a go-to method for process improvement.

We'll come on to why this is problematic in a minute. But the good news is, this number is finally decreasing as businesses transition away from process mapping to pursue process mining instead. They've seen significantly higher returns being generated across multiple industries from Retail and Life Sciences to Banking and Manufacturing.

So what's the deal? What exactly is the difference between process mapping and process mining, and is one truly better than the other?

What is process mapping?

The short answer is that it's exactly what it sounds like — a visual representation of how work happens, often in the form of a process flowchart or value stream map.

It's helpful for bringing visibility to the individual components of an existing process or workflow, and can help guide basic decision making for core processes, such as in the example below. Most process maps make use of a common language made up of symbols that everyone in the organization can easily understand.

Now for the longer answer on what process mapping is: Although a helpful starting point for process improvement, for most businesses, it's also a pretty time-consuming and labor-intensive exercise that yields limited results compared to the amount of effort put in. For example, interviews with process owners, managers and other stakeholders are required to get a better understanding of the steps or tasks involved in each process. The outputs of those interviews then get translated into whiteboards and sticky notes (or the digital equivalent of sticky notes…which isn't much faster).

Even if those initial steps go smoothly, they come with the significant caveat that the opinions gathered at the interview stage may be biased, incomplete or incorrect, and the processes mapped to them will only represent snapshots of reality, rather than what's unfolding in real time. Plus, by the time the last sticky note hits the wall, an entire process may have changed, given the entire exercise can take months.

After that, there's the challenge of identifying the pain points within a process map (which, given the above caveats, might not even be an accurate representation of how work is happening), followed by a reordering or replacing of individual tasks or process steps, and finally, a whole new round of map drawing.

Fortunately, business process mapping has evolved somewhat in recent years, allowing for integrations between process mapping software and business process management systems, but this still leaves businesses in a position where they have to manage their complex processes across fragmented technologies, while relying on potentially misleading or incorrect information.

So is process mining just process mapping on steroids?

It's easy to see why this assumption gets made, as one of the core outputs of process mining software is a set of detailed process maps. But process mapping and process mining couldn't be more different.

Unlike business process mapping software, process mining technology takes real-world data from time-stamped event logs (and objects, if you're using object-centric process mining), such as when a purchase order (P.O) was created, approved, fulfilled, and dispatched. It then creates an objective and complete view of the processes taking place at scale across your organization, as they're happening. Machine learning is then applied to create immediate, self-service observations and recommendations, while corrective actions are automated so they can be executed in real time.

In other words, instead of showing you what people think work looks like, process mining shows you what's really happening beneath the surface. You can think of it as an M.R.I of your business processes, giving you a living, breathing, moving picture of how your processes actually run, and unearthing high-value improvement opportunities in real time.

Diving deeper, process mining consists of the following five steps:

Data ingestion: During this step, event data is pulled from the information systems that run your business operations, such as your E.R.P, C.R.M and S.C.M, to create an unbiased view of what's happening across your organization.

Process discovery: This is where process mining technology creates an end-to-end visualization — basically a live, highly detailed process map — of your workflows, revealing insights such as where slightly different process steps were taken in real life, known as “variants,” or which variants caused the process to take longer. It would be fair to say that thisstep alone is a bit like process mapping on steroids...

Process analytics: This is the step where process improvement opportunities are revealed across your organization and quantified across your K.P.I's. Crucially, this is the step that also separates process mining from workflow process mapping software and basic R.P.A bots, as it takes process analytics a step further to show you the root causes of missed K.P.I's.

Process benchmarking: This step gives you the ability to compare process performance across different dimensions, such as how long it takes to complete a process or task in one country versus another, or when using a particular supplier versus others.

Conformance checking: This step reveals commonalities and discrepancies between modeled behavior and observed behavior. In other words, it tells you whether particular tasks or process steps are being skipped, prolonged or executed in the wrong order.

It's a lot to take in, but hopefully the takeaway is clear: Process mining isn't remotely like process mapping on steroids. It's an entirely different beast that facilitates continuous improvement, generating a significantly higher R.O.I.

All of this adds context to Gartner's prediction that by 2025, 80% of organizations driven by the expectations of cost reduction and automation-derived enhanced process efficiency will embed process mining capabilities in at least 10% of their business operations.

Is process mapping easier to achieve than process mining?

The answer to this question may seem like an obvious “yes” as it's undoubtedly easier to draw a process flowchart or value stream map than to complete all the above steps. But that's a false economy — especially if you end up with an inaccurate, out-of-date process map with no meaningful actions automated as a result, or if you're spending millions with consultants to do it.

Plus, if you conduct a thorough process mining vendor evaluation, you should be able to find one who can provide all of the above process mining capabilities for you. For example, if you decide to take an platform approach to process mining (which is where process mining is embedded into a wider platform to continuously manage and orchestrate your business processes), your process mining platform would automatically create objective process maps for you, while running all the other steps in tandem.

Process mapping vs process mining: Which is the best solution for you?

Ultimately, it's up to you and your stakeholders whether process mapping or process mining makes the most sense for your business right now. Diving deeper into the specific use cases you're trying to solve is probably the best next step for figuring that out.

But hopefully one thing is clear: Process mapping and process mining couldn't be more different. Traditional process mapping was a longstanding cornerstone of business process management (100 years isn't a bad run…), and it has earned its place in B.P.M history. But if you're looking for a process improvement solution that'll get your business working faster and more effectively both now and in the future, process mining could be for you.

Process Intelligence vs Process Mining: What's the difference?

Process mining and process intelligence could, at first glance, be mistaken as interchangeable. Understandable, since both give you visibility over how your organization's processes run. Process Intelligence is probably just a fancier name for process mining, you might think. But let us stop you there.

Process mining walked so Process Intelligence could run. All the work process mining tools do to create business-wide visibility has been something of a pathbreaker for Process Intelligence. So there are some simple but important differences to clear up, giving you a full understanding of the sophistication of Process Intelligence and the value it can offer your business.

Let's start with the concept you're probably more familiar with.

What is process mining?

Your familiarity with process mining is a safe bet – you're on the Celonis website, after all. Perhaps you've bumped into our 'For Dummies' guide. And we also know that 70% of businesses are either already using process mining or exploring using it.

Here's a quick summary for anyone brand new. Process mining is a widely-used technology to model, analyze, and optimize business processes. It provides an objective view of how processes run across systems, desktops, and departments. In doing so it helps teams improve processes that were originally designed to be linear but soon ended up like a bowl of spaghetti. Improvement does not only mean visibility, it includes enabling people to find and capture value in processes.

What does process mining do?

Now for the nitty gritty. Process mining uses data from your systems of record in the same way task mining collects desktop data from activities such as sending emails and opening spreadsheets. Think E.R.P's, C.R.M's, S.C.M's – basically anywhere you have data related to the running of your business.

Process mining therefore accounts for all the variants in your processes – from procurement to fulfillment. But the important thing to say here is that it's real-time data. This means the process modeling isn't just highly detailed, but lives and moves with your business.

Process mining is great for getting complete visibility over your processes, so you can begin to prioritize where to take action and how. It's also one of the core technologies that powers Process Intelligence and feeds into the Celonis Context Model, which enables you to improve processes and boost company performance.

What is Process Intelligence?

Process Intelligence is the tissue that connects the different parts of the business—departments, systems, and people. Sitting at the heart of the Celonis Platform within the Celonis Context Model, it gives you a common, system-agnostic language for how your business actually runs. In doing so, it gives you a common language for how the business really runs, across every system and each and every department. Because it's system agnostic and unbiased you get an objective view – the data doesn't lie.

Teams use Process Intelligence to build on the visibility that process mining provides by enabling you to take easy, effective action. It's here where the exciting stuff happens for process optimization and value realization. With informed decisions, you can effectively target and capture the value hiding in processes.

What does Process Intelligence do?

Process Intelligence gives you the power to see your processes and improve them. It goes beyond process mining to connect you to your processes, your teams to each other, and emerging technologies to your business fast. That's why we call it the connective tissue of the enterprise.

It takes data from systems like your E.R.P's, C.R.M's, and Excel, combining process data with business knowledge into the Celonis Context Model—a living, system-agnostic digital twin of your end-to-end operations.

This is where the Celonis Context Model takes you further. By adding a dynamic intelligence layer to your process data and business knowledge, it provides hindsight, insight, and foresight (predictions, simulations, and recommendations via Decision Intelligence). This enables Enterprise A.I to reason correctly and take trusted action.

Finally, Process Intelligence gives you the tools to capture that value. For instance by empowering your people with easy-to-action insights, and by helping you effectively deploy technologies like A.I and automation to improve your processes.

Being system-agnostic, it all works with what you currently use, without the worry of incompatibility or replacement. So the short answer is: it can do an awful lot.

At Celonis, we've seen Process Intelligence create billion-dollar free cash flow improvements by optimizing cash collection effectiveness; massive bottom-line and green-line savings thanks to reducing the distance traveled by delivery trucks; and double-digit reductions in cancellations by optimizing order fulfillment cycle times.

Why is Process Intelligence important?

Process Intelligence is a new class of intelligence that helps you get more R.O.I from your existing technologies and deploy new ones more effectively. Through the Celonis Context Model, Process Intelligence sits in a dynamic context layer between your data and your A.I applications. It feeds A.I agents the real-time operational context they need to execute accurately, prevent costly hallucinations, and maximize tech stack R.O.I.

By introducing a shared language for your departments, systems, and processes, Process Intelligence creates a shared understanding for achieving organizational alignment and identifying value opportunities.

Process Intelligence versus Business Intelligence: What's the difference and which problems do they solve?

Business Intelligence and Process Intelligence can be easily confused. Both offer data-driven approaches to answering questions about your business' performance. Both promise data-based insights, increased efficiency, better customer experience and empowered employees.

Some purists will claim one is much better than the other, but in reality, both have their purpose.

So, here's a breakdown of Business Intelligence versus Process Intelligence, their similarities, differences, and which problems they can solve for your company.

What is Business Intelligence?

Business Intelligence (B.I) is an umbrella term that refers to various tools, methods, and processes for collecting, evaluating, and presenting data about a business. Companies use Business Intelligence to understand and measure performance against their goals. B.I combines data from different sources and presents it in user-friendly visualizations like dashboards, graphs, charts, and entire reports. Users can analyze this information to gain insights into patterns and discontinuities and answer questions about their company, competitors, customers, or market developments.

Historically, B.I was mostly in the domain of data analysts – they analyzed the data and created reports for the wider organization. In contrast, modern B.I solutions have a strong focus on self-service, so employees can ask and answer questions even without a technical background, and share data-driven insights with their colleagues and stakeholders. However, even with comprehensive self-service capability, many organizations will still require help from other departments to maximize results. For example, the I.T department will still be needed to move data from the source systems into the B.I tool of choice and to "deepen" the data analysis.

How does Business Intelligence work?

In very general terms, B.I tools use raw data collected from business systems, for example databases, reports, and spreadsheets, but also email or social media. This data is processed and then visualized for further analysis and querying.

B.I is designed to answer specific questions and provide snapshot analysis for decision making or planning by using a wide range of methods and tools, such as data preparation,data visualization, reporting, benchmarking, or querying. All these tools help users to understand business performance, follow trends, and make better decisions.

What problems does B.I solve?

It's important to understand that B.I typically focuses on what is happening. It covers what is called descriptive analytics – telling you what's currently happening inside your business, and everything in the lead-up to now. For example, B.I can help you answer questions like:

What are our best performing products by country?

How has that changed over the last three months?

How are we performing in comparison to the same time last year?

Are our customers paying us on time?

Is there anything going wrong that we need to change?

But helpful as this is, B.I doesn't tell you why or how these things happen, nor what actions you should take. So, you might very clearly see that you've got a problem, but to fix it, you need more info — like exactly where the problem is coming from, the parties involved, and the K.P.I's and other processes it's affecting. That's where Process Intelligence comes in.

What is Process Intelligence?

Your business runs on processes. They are at the very core of how things happen inside your company. Your supply chain, Purchase-to-Pay, Order-to-Cash, or customer journey – these are all processes that drive operations. But most companies only have a very rudimentary understanding of how their processes really work, how they impact each other, and how many ways there are to complete a process.

Even the people working within a process usually only have a limited understanding beyond their part in it. And in a worst-case scenario, the different departments inside a company speak completely different languages when it comes to understanding goals and dependencies.

Process Intelligence aims to provide that much-needed shared understanding, as well as clarity on how processes run and where optimization opportunities are hiding. It's a set of tools and methods to create end-to-end transparency about the processes and workflows that run in an organization, like the steps and parties involved, dependencies between processes, plus up and downstream effects. Crucially, it also comes with the technologies to optimize processes, unlock hidden value, and enhance company performance.

How does Process Intelligence work?

Like B.I, Process Intelligence analyzes and visualizes data from various sources. The difference is that Process Intelligence looks at information about how processes run, interact, and depend on each other continuously.

Users can on one hand see how their processes truly run to understand variations and spot problems as well as root causes of bottlenecks and rework. On the other, they can rely on various Process Intelligence tools and features that allow them to go beyond analysis and actually solve problems within a process.

Process Intelligence thus includes several technologies and methods for both analysis and execution, such as process mining, process modeling, task mining, the digital twin of an organization, simulation, monitoring, automation, and generative A.I.

What problems does Process Intelligence solve?

Where B.I asks what is happening, Process Intelligence asks how and why it's happening, and what can be done to make things better. You can think of it as the connective tissue between people, processes, and technology, providing everyone in your organization with a common language for how your business is running, visibility into where value is hiding, and the ability to capture it.

A fully running Process Intelligence solution can not only tell you where you can lower costs, how you can speed up delivery, or what you need to do to speed up collections. It can also tell you what will have the biggest impact and take the lowest effort, so you can prioritize accordingly. You can also use it to set up automatic workflows, for example to unblock orders, or set up alerts to your teams when K.P.I's fall out of range.

With the recent addition of generative A.I to the Process Intelligence tech stack, your employees can now directly ask questions in a language they understand to solve process problems. And they can share these insights across departments to create a common understanding of dependencies and priorities.

Process Intelligence versus Business Intelligence – what's the difference?

To summarize – Process Intelligence primarily focuses on the analysis and optimization of business processes such as Purchase-to-Pay, supply chain management, or customer journey to increase business performance. It offers businesses advanced capabilities like root cause analysis, workflow automation, real-time monitoring, alerts, and process optimization. In simple terms, it tells you how and why things happen and most importantly, what steps you should take to unlock value opportunities hiding inside your processes.

Business Intelligence on the other hand is concerned with providing an at-a-glance view of an organization's performance, market trends, and customer behavior. It tells you what's happening inside your organization and how you got there, using descriptive analytics, performance benchmarking, and reporting.

Both make use of Machine Learning and A.I to ask questions and drill deeper into the respective data.

So, as you can see, B.I and Process Intelligence both have their purpose. Depending on what you want to do, you can use one or the other or a combination of both.

Celonis Process Intelligence Graph provides a common language for enterprise process performance

Data by itself has no meaning, no value, like ancient hieroglyphics you can't read. But add context, a way to translate or understand the data, and suddenly meaningless information becomes knowledge. Apply that knowledge to a practical purpose, and then you have something truly special – intelligence. Today, Celonis announced a new innovation in process intelligence, the Process Intelligence Graph (P.I Graph).

For more than a decade, companies have used Celonis technology to collect and analyze process data. Originally, looking at a single process (e.g., procurement) or object (e.g., purchase order) at a time. Now, thanks to the Object-Centric Data Model, all an organization's process data can be unified into a single source of truth, a system-agnostic, digital twin.

With the twin (i.e., the data) in place, the next step is adding the context through the Celonis Process Knowledge Layer. Announced at the event, the Process Knowledge Layer is a new capability of the Celonis platform that enables customers to understand why their twin looks the way it does, where the value opportunities lie and how to deliver on them. This knowledge applied to real-world company data produces process intelligence – the P.I Graph.

“The Process Intelligence Graph is the modern equivalent of the Rosetta stone - it's the connective tissue that's been missing in modern enterprises,” said Alex Rinke, co-C.E.O and co-founder of Celonis, in a press release. “By bringing together the building blocks of data, knowledge, and the power and scale of our ecosystem, our customers now have the foundation they need to make processes and entire value chains work for everyone. This level of cross-process, cross-system intelligence can only be achieved with the Celonis platform and the Process Intelligence Graph at its heart. The P.I Graph enables technologies like A.I and automation, feeding them the data they need for how processes actually run, contextualized with the knowledge of why they run the way they do, and how they can be improved.”

The Process Intelligence Graph was one of multiple innovations and new platform capabilities Celonis announced at Celosphere 2023, its annual user conference in Munich. During the two-day event, the company also debuted Process Copilot, Transformation Hub, Celonis Studio updates and a new Material Emissions App.

Celonis Process Intelligence Graph (P.I Graph)

The Process Intelligence Graph is designed to help companies realize more value at greater speed. It enables new technologies, like artificial intelligence. And, it is the foundation to optimizing not just processes, but entire value chains.

During the keynote, Divya Krishnan, V.P of product marketing at Celonis, described how the P.I Graph speeds up value realization with Celonis, making it faster to extract transactional data, find the right use case, and actually capture value.

“First, you can model your process data faster. Deploy once and scale infinitely from there,” she said. “Next our process knowledge serves up recommended, prioritized opportunities for you to look at. Finally, you can use prebuilt apps to capture value fast.”

These apps, Krishnan said, are system agnostic and can be used on top of any system, whether it's Infor, sap, Oracle or even homegrown systems.

The P.I Graph's out-of-the-box definitions of objects and events means teams need less S.Q.L knowledge to model process data and that the models can be reused for future processes, allowing customers to deploy once and scale infinitely. Additionally, the P.I Graph can apply standardized, out-of-the-box process knowledge to customer data, providing recommended, prioritized opportunities.

The P.I Graph helps organizations to:

Generate automated, actionable insights across processes, systems and departments

Identify and remove replicated processes faster

Build applications that are enhanced by the entire ecosystem

Deliver a better customer experience, backed by reduced emissions and more sustainable supply chains

Make Process Intelligence an imperative layer in their enterprise technology stack for A.I to work most effectively

The P.I Graph is also being constantly enriched and extended with external partner data, such as carrier information, and process knowledge on occurrences like high safety stocks impacting working capital; and machine learning technology and models that are unique to process mining on objects that can cause bottlenecks, such as duplicate invoices, late orders, or mismatched terms.

At the heart of the P.I Graph are the Object-Centric Data Model and Process Knowledge Layer:

Object-Centric Data Model: Launched during Celonis World Tour 2023, the Object-Centric Data Model (O.C.D.M) is an extensible data representation of an entire business. It contains object types, event types, object-to-object relationships, and event-to-object relationships, built and arranged to model what happens in the supported business processes. The O.C.D.M is the core of an organization's digital twin. The O.C.D.M enables object-centric process mining innovations across the Celonis platform. For example, it powers the Process Explorer and End-to-End Lead Times App.

Process Knowledge Layer: The Celonis Process Knowledge Layer, announced at Celosphere 2023, acts as a centralized source of truth for process knowledge. It enables users to explore and maintain process knowledge from one place. It also automatically enriches data with additional knowledge powered by Machine Learning. This makes it easier to effectively drive insights and value realization with tools across the Celonis platform

Celonis Object-Centric Data Model: Single source of truth for process intelligence

Object-centric process mining is a true revolution in process mining technology. Celonis fully embraced object-centric process mining with the release of Process Sphere. Today, at the first stop on the Celonis World Tour 2023, which took place in Munich, Germany, the company followed up by announcing enhancements for Process Sphere and releasing a major, new O.C.P.M innovation, called the Object-Centric Data Model.

“Object-centric process mining allows users to easily navigate through processes based on real-life objects and events, and fully unleashes the power of process mining,” said Alex Rinke, co-founder and co-C.E.O at Celonis, in a press release. “Customers can build and interact with their data more naturally, in a way they are already familiar with.”

Along with the Object-Centric Data Model and Process Sphere updates, Celonis announced two additional O.C.P.M innovations, the Multi-Object Process Explorer and End-to-End Lead Times App. Celonis also showed a beta of L.L.M for P.Q.L Generation, a forthcoming A.I capability for Celonis E.M.S that translates natural language user queries into Process Query Language (P.Q.L).

The Object-Centric Data Model is generally available now, the Process Sphere enhancements will enter general availability in the Fall.

Object-Centric Data Model

Object-centric process mining (O.C.P.M) overcomes the limitations of traditional process mining techniques and allows organizations to better visualize and analyze the complexity and interconnectedness of modern business operations.

Process Sphere is a capability of Celonis E.M.S that uses O.C.P.M to provide end-to-end visibility of business operations. Process Sphere allows businesses to visualize and analyze the relationships between objects and events across interconnected processes. For example, Process Sphere can show you if procurement problems are affecting sales order fulfillment.

The Process Sphere enhancements will enable customers to more easily explore multi-object processes and relationships and improve bottleneck detection. Analysts will also see recommended prompts and suggested use cases that they can use to structure their analysis, and will be able to better assess process conformance and deviations against process models.

The Object-Centric Data Model is an extensible data representation of an entire business. It acts as a single source of truth for all process intelligence and serves as the core of an organization's digital twin. The model reduces the work needed to transform data from source systems (e.g., E.R.P, C.R.M, S.C.M, etcetera) and operates side-by-side with existing Celonis E.M.S Event Log Data Models, making the transition for existing customers non-disruptive. Customers can continue using their current data models for existing processes, and take advantage of the new model for high-value use cases that require cross-process analysis.

The benefits of the Object-Centric Data Model include:

Greater modeling simplicity: Businesses can work with a data model that uses the same language they do–for invoices, orders, deliveries–rather than the language of their source systems, to better model the business. Additionally, the data model objects are reusable, meaning businesses no longer need to reinvent the data pipeline for each new project.

Faster, system-agnostic implementations: Companies can use Celonis' apps and process content, no matter which vendor's source system they use, leveraging standardized business definitions and prebuilt transformations for core processes.

More analytic flexibility: With traditional process mining, an organization must create a data model for each process it wants to analyze. With the O.C.D.M, organizations can dynamically adjust process analyses, switching perspectives from process to process without needing to go back to the source data.

Multi-Object Process Explorer and End-to-End Lead times App

The Object-Centric Data Model enables object-centric process mining innovations across the Celonis platform. For example, it powers the new Multi-Object Process Explorer and End-to-End Lead Times App.

The Multi-Object Process Explorer is an enhanced version of the existing Celonis Process Explorer, which allows users to explore how process activities are connected. The new Multi-Object Process Explorer lets customers see how objects relate to each other and how an event relates to multiple objects involved in a process, something not possible with traditional process mining. Customers can use the new tool for initial exploratory analysis and from there, create custom views that address complex multi-object relationships in the Celonis Studio.

The End-to-End Lead Times App provides supply chain transparency and enables customers to reveal and mitigate bottlenecks in the creation of finished goods from sourcing raw material to final delivery. With this information, the company said, supply leaders can act more accurately and quickly to “improve service levels and accelerate cash conversion through better working capital.”

What is a process digital twin? The answer is evolving due to process mining

The future of digital twins is evolving to be more process driven. Enter the process digital twin.

What is a process digital twin?

Typically, digital twins are defined in the context of the Internet of Things, industrial applications, and physical assets such as aircraft engines, turbines and manufacturing equipment.

Per Gartner:

A digital twin is a digital representation of a real-world entity or system. The implementation of a digital twin is an encapsulated software object or model that mirrors a unique physical object, process, organization, person or other abstraction. Data from multiple digital twins can be aggregated for a composite view across a number of real-world entities, such as a power plant or a city, and their related processes.

To scale, digital twins are going to become more about systems and processes that revolve around financials, supply chain, manufacturing and logistics. The next rev of digital twins will revolve around the real world simulation of businesses and how they operate instead of physical assets.

I caught up with a Celonis customer using digital twins for multiple processes. The idea is to use Celonis to replicate their business operations in digital form, leverage analytics and real time data and run simulations to enable digital transformation.

What was notable about this customer's approach is that digital twins were being built in conjunction with process mining. The data coursing through cloud applications and Celonis were being used to simulate automation flows and end-to-end processes. A process digital twin will be able to see how processes actually work within a business and going forward these models can find interdependencies across units and partners.

In other words, with Celonis, every transactional event and global business service can have a digital twin.

For now, process digital twins are being implemented at companies with value being realized. Future efforts will revolve around operationalizing process digital twins and then integrating them into a broader automation, analytics and process excellence effort. "Digital twins of people, processes, organizations and environments will be used for strategic and operational decision making and advanced simulation," according to Gartner.

Use cases for process digital twins

These process digital twins will also be relevant to a bevy of horizontal use cases, including:

Accounts Payable;

Accounts Receivable;

Procurement;

Order Management;

Inventory Management;

Supply Chain resiliency, which is increasingly a boardroom issue;

And multiple shared services.

The prerequisite for process digital twins will be cloud architecture and the systems transformation required to tap into multiple systems and real-time data. These process digital twins have also been called Digital Twin of an Organization (D.T.O), a term popularized by Gartner. The research firm said in a July 2021 report:

Gartner expects that D.T.O's will become critical as digital business systems are increasingly reliant on continuous integration of human and machine intelligence. A D.T.O reflects this real-world environment with real people and machines working together, and allows users to model different scenarios, choose one and then make it real in the physical world.

Today, those D.T.O's are edging closer to reality from the vision presented in 2019.

With process digital twins evolving it's worth highlighting the differences and similarities with more physical-based digital twins.

What is a process digital twin or D.T.O versus physical object-based digital twin?

By definition, digital twins cover both physical assets and processes, but the deployments to date have been about representing real-world systems. Physical digital twins have data based on physics, maintenance, operations and simulations. Process digital twins are focused on understanding the process. The digital footprints that reside in systems of record are what drives process digital twins.

A process digital twin may have the following characteristics:

Data is primarily generated through transaction processing instead of physical and real-world assets and sensors attached to them.

Value is created through speed, efficiency and cash flow instead of opportunities like preventative maintenance.

I.T systems provide the data used to create the digital twin and run simulations. A digital twin that replicates a physical asset is built with sensor data.

Automated and human work are also combined with process digital twins. A process digital twin must account for end-to-end processes and simulate how automation and human work interact. For instance, automation early in a process may lead to human work bottlenecks later.

In other words, process digital twins have a few moving parts.

Celonis Chief Scientist Wil van der Aalst recently co-authored a research paper on the process-driven Digital Twin of an Organization (D.T.O) concept. The paper noted:

Why is it so challenging to create a D.T.O? There are two main reasons:

The boundaries of an organization and, therefore, also a D.T.O are not so clear, that is, an organization has customers, suppliers, employees that collectively influence the processes.

Human and organizational behavior may be irrational and change over time (influenced by regulations, social interactions, and personal preferences).

For most organizations, it is not feasible to create a D.T.O that captures reality well. However, the desire to model, visualize and understand the complex context in which an organization operates is compelling. One can view process mining as a concrete technology to facilitate such a D.T.O. Using process discovery, one can discover the so-called"control-flow model" (represented using pee-tree nets, process trees, or B.P.M.N models) and by aligning event data with the control-flow model, it is possible to add other perspectives (time, costs, resources, decisions, etcetera).

Van der Aalst outlined how D.T.O's are more vision than reality, but innovative enterprises are getting closer with process mining. For instance, CapGemini is using Celonis to create digital twin replicas of businesses.

And the core areas of D.T.O's are also maturing. Gartner outlines five pillars of D.T.O's.

Digital optimization, optimized business operations, improved business model or digital business transformation.

Digitalized business operating model.

Business performance management frameworks to connect models and measurement.

Business operations intelligence to provide the model with real-time data.

Value for the different stakeholders (internal/external) of the organization.

Celonis Process Intelligence Graph provides a common language for enterprise process performance

Data by itself has no meaning, no value, like ancient hieroglyphics you can't read. But add context, a way to translate or understand the data, and suddenly meaningless information becomes knowledge. Apply that knowledge to a practical purpose, and then you have something truly special – intelligence. Today, Celonis announced a new innovation in process intelligence, the Process Intelligence Graph (P.I Graph).

For more than a decade, companies have used Celonis technology to collect and analyze process data. Originally, looking at a single process (e.g., procurement) or object (e.g., purchase order) at a time. Now, thanks to the Object-Centric Data Model, all an organization's process data can be unified into a single source of truth, a system-agnostic, digital twin.

With the twin (i.e., the data) in place, the next step is adding the context through the Celonis Process Knowledge Layer. Announced at the event, the Process Knowledge Layer is a new capability of the Celonis platform that enables customers to understand why their twin looks the way it does, where the value opportunities lie and how to deliver on them. This knowledge applied to real-world company data produces process intelligence – the P.I Graph.

“The Process Intelligence Graph is the modern equivalent of the Rosetta stone - it's the connective tissue that's been missing in modern enterprises,” said Alex Rinke, co-C.E.O and co-founder of Celonis, in a press release. “By bringing together the building blocks of data, knowledge, and the power and scale of our ecosystem, our customers now have the foundation they need to make processes and entire value chains work for everyone. This level of cross-process, cross-system intelligence can only be achieved with the Celonis platform and the Process Intelligence Graph at its heart. The P.I Graph enables technologies like A.I and automation, feeding them the data they need for how processes actually run, contextualized with the knowledge of why they run the way they do, and how they can be improved.”

The Process Intelligence Graph was one of multiple innovations and new platform capabilities Celonis announced at Celosphere 2023, its annual user conference in Munich. During the two-day event, the company also debuted Process Copilot, Transformation Hub, Celonis Studio updates and a new Material Emissions App.

Celonis Process Intelligence Graph (P.I Graph)

The Process Intelligence Graph is designed to help companies realize more value at greater speed. It enables new technologies, like artificial intelligence. And, it is the foundation to optimizing not just processes, but entire value chains.

During the keynote, Divya Krishnan, V.P of product marketing at Celonis, described how the P.I Graph speeds up value realization with Celonis, making it faster to extract transactional data, find the right use case, and actually capture value.

“First, you can model your process data faster. Deploy once and scale infinitely from there,” she said. “Next our process knowledge serves up recommended, prioritized opportunities for you to look at. Finally, you can use prebuilt apps to capture value fast.”

These apps, Krishnan said, are system agnostic and can be used on top of any system, whether it's Infor, sap, Oracle or even homegrown systems.

The P.I Graph's out-of-the-box definitions of objects and events means teams need less S.Q.L knowledge to model process data and that the models can be reused for future processes, allowing customers to deploy once and scale infinitely. Additionally, the P.I Graph can apply standardized, out-of-the-box process knowledge to customer data, providing recommended, prioritized opportunities.

The P.I Graph helps organizations to:

Generate automated, actionable insights across processes, systems and departments

Identify and remove replicated processes faster

Build applications that are enhanced by the entire ecosystem

Deliver a better customer experience, backed by reduced emissions and more sustainable supply chains

Make Process Intelligence an imperative layer in their enterprise technology stack for A.I to work most effectively

The P.I Graph is also being constantly enriched and extended with external partner data, such as carrier information, and process knowledge on occurrences like high safety stocks impacting working capital; and machine learning technology and models that are unique to process mining on objects that can cause bottlenecks, such as duplicate invoices, late orders, or mismatched terms.

At the heart of the P.I Graph are the Object-Centric Data Model and Process Knowledge Layer:

Object-Centric Data Model: Launched during Celonis World Tour 2023, the Object-Centric Data Model (O.C.D.M) is an extensible data representation of an entire business. It contains object types, event types, object-to-object relationships, and event-to-object relationships, built and arranged to model what happens in the supported business processes. The O.C.D.M is the core of an organization's digital twin. The O.C.D.M enables object-centric process mining innovations across the Celonis platform. For example, it powers the Process Explorer and End-to-End Lead Times App.

Process Knowledge Layer: The Celonis Process Knowledge Layer, announced at Celosphere 2023, acts as a centralized source of truth for process knowledge. It enables users to explore and maintain process knowledge from one place. It also automatically enriches data with additional knowledge powered by Machine Learning. This makes it easier to effectively drive insights and value realization with tools across the Celonis platform.

Celonis Platform One platform. Transformational impact.

The trusted platform to industrialize Enterprise A.I

The Celonis Platform brings together process data, business knowledge, and intelligence from all of your systems, applications, and devices to create a dynamic, real-time digital twin of your operations.

It combines hindsight, insight, and foresight to give your people and your A.I agents the operational clarity to reason correctly, decide sensibly, and act reliably.

Then, the Platform provides you with powerful capabilities to analyze, design and operate A.I-driven processes.

These capabilities are built to support the full enterprise A.I lifecycle, enabling you to effectively integrate A.I into your operations.

Image summary: A conceptual architectural diagram on a black background depicting a three-layered system stack. The top layer, labeled BUILD EXPERIENCE, is surrounded by various functional icons. The middle layer is labeled CONTEXT MODEL and features a circular arrangement of small blocks. The bottom layer is labeled DATA CORE and consists of a grid-like structure. Each of the three layers is linked to an API on the right side.

The Data Core

The Celonis Platform is made up of three layers. First, the Data Core, our high-performance data infrastructure.

Data Core lets you extract data from any source and query billions of records at speed, scaling to even the largest enterprise volumes.

Data Core

The Context Model

The Context Model

The Celonis Context Model is the heart of the Celonis Platform. It creates a dynamic, system-agnostic, real-time representation of your entire business.

The Context Model is built on process data and business knowledge from across systems, applications, devices, and interactions. Based on this data and knowledge, the C.C.M provides intelligence to determine root causes, generate predictions and recommendations, evaluate what-if scenarios, and perform other analyses and calculations.

Powered by this Context Model, your people and your Enterprise A.I have the operational clarity needed to reason correctly, decide sensibly, and act reliably, producing meaningful business impact.

The Context Model is continuously improving, mining new processes and agents as they execute, and learning from how your people and agents interact in real time, constantly evolving with your business.

The Build Experience

Analyze, design, and operate composable, A.I-driven business processes. Build and run strategic, operational, business-critical solutions.

Analyze

Understand how your processes truly run. Identify the most impactful opportunities to implement improvements and deploy A.I. Run process simulations, predictions, and what-if scenarios.

Design

Redesign operations to integrate Enterprise A.I. Re-engineer processes based on deep process insight. Define workflows, outcomes, guardrails, and best practices.

Operate

Continuously monitor process performance, adherence, and agent activity. Orchestrate A.I to work alongside people and systems.

You have reached the end of the document.