The Year Ahead for BPM -- 2019 Predictions from Top Influencers
- Published: January 22, 2019
- Written by BPM
It’s all about automation - this is an interesting shift, since just a couple of years ago a good number of BPM experts thought it’s all about adaptive case management and how to support knowledge workers. The shocking reality however is that most companies don’t want to support knowledge workers - they want to replace them.
Now we’re seeing AI dominating the conversations (without a lot of actual adoption in the early majority), and RPA creating a lot of buzz, though companies adopting it are starting to realize that scaling an RPA based automation architecture is flawed by design.
In 2019, I expect RPA to plunge into the trough of disillusionment, though the appetite for automating business processes is far from satisfied. Leading analyst firms seem to return to the roots of BPM as an orchestration layer implemented by workflow automation, and at least from where we stand it looks like this is a rapidly growing, sustainable market.
- The current Artificial Intelligence (AI) fascination is unfortunately completely biased on Deep Neural Networks (DNN) and Machine Learning (ML) for everything. Regulated and unregulated industries should advance with caution. Explainable AI is a must for automated business decisions.
- Decision Management, and DMN (Decision Model and Notation) in particular, is now recognized as the next best tool in Business Process Management (BPM). Allowing organizations to inject knowledge-based decisions services that are traceable, auditable and explainable into the process fabric of their operations. DMN will progressively replace traditional Business Rule Management Systems (BRMS).
- The microservice movement will reignite the need for orchestration. Some analyst is bound to rename BPM engines to Microservice Orchestration Engines (MOE)..wait did I just do that.
- BPM as a practice will continue to shift its interest from efficiency of operations (OPEX) to digital transformation (DX), shifting the organizational focus from Inside-Out to Outside-In value creation. Continuing the journey toward true business agility initiated in the software realm.
- Business Architecture is growing as a movement, but it will only find success if it is able to provide an agile method for business transformation. Too many still believe that the architecture is the deliverable, when the real desired outcome is agile business transformation.
From my perspective BPM has stagnated in the last few years. Yes, the pundits jumped on every new buzzword that came around and that includes AI — when there is no such thing — but underneath it all stayed the same. BPM is still a methodology and some software and the arguments as to why are still more agility, higher quality, and less cost. All of them virtually not true. So BPM is today another form of low-code application development.
If there is one thing that BPM has to do is to abandon the idea that a business process is a standardised and reusable business asset. It will be wrong for most of its users. The asset is the understanding of what the process delivers in outcome … circling back to goal orientation. Mapping outcomes in a value stream to the organisation that will deliver it to which customers … that is BPM. BPM must start to focus not on controlling humans but on augmenting them. That augmentation must be in a form attractive to humans while enabling security, compliance, authenticity and auditability. And herein lies the true challenge!'
Customer experience will become the North Star for all value-added operational excellence and business transformation initiatives. Successful organizations will differentiate themselves by ensuring the customer experience is not a fashion or an afterthought, but instead lies at the very heart of how they organize and run their business.
2019 will be the beginning of the end for stand-alone process mining, with the market likely to entirely disappear by the end of 2020.
Savings can be made in analyzing key end-to-end system processes, but organizations are questioning the high initial costs. Process mining will be incorporated into larger suites of professional business process analysis tools. This will aid the faster creation of baseline models, enabling analysis and process improvement.
The continual monitoring of process health will lead to a switch in the way process mining is applied, to real-time process monitoring. Leveraging process mining technology in this way will ensure that organizations can more readily obtain cross-system views.
The inability to scale Robotic Process Automation (RPA) implementations will reduce expected return on investment
The majority of RPA rollouts remain at the project level, with many large organizations deploying or testing different RPA tools, vendors and solutions. This fragmentation could deal a killer blow to the promise of large-scale adoption: A Gartner survey found 24% of respondents cited scaling RPA as their number one problem.
In early RPA implementations, many vendors shunned the need to partner or become involved in a broader solution. In 2019, RPA vendors will realize upsells aren’t so simple—better planning and support is needed. Enterprises must consider where and when to apply RPA, to achieve the expected value returns.
BIG CHANGE will be the dominant process theme in 2019. Organizations of all sizes will finally realize that incremental improvement, aka “better sameness” won’t save them from the existential threat of digital disruptive competitors, Only by launching large-scale process projects can traditional, industrial-era organizations shed their suffocating legacy of hierarchy and survive. Without BIG CHANGES, many risk-averse organizations will slowly or rapidly disappear into self-imposed oblivion.
The upside is that BIG CHANGE can produce HUGE BENEFIT when analog old process designs become transformed into agile digital beauties. Those organizations that can figure out how to launch, execute, and sustain BIG CHANGE will the big winners in 2019.
Top 5 List of what to expect in the BPM market in 2019:
- Event-Driven – The future of BPM is event-driven. As we move into a world that is more and more dominated by technologies such as big data, IoT, and ML, more and more processes will be started by external events. This requires a new architecture capable of true scale and speed.
- hpaPaas vs. BPM vs. Low Code – Gartner has confused the markets with its acronyms. In 2019 Gartner will get sensible and merge terms into just the Low Code market with some differentiation between model driven vs. code generation execution and between process-centric vs. data-centric app building.
- ML/AI – The AI we see on our favorite Netflix sci-fi is still a few decades away. However, ML will begin to drive real value in BPM process models by eliminating clunky OCR, powering chatbots, and improving decision rules.
- Solutions not Products – Customers are tired of buying BPM bloatware. Customer demand for solutions built on lean, loosely coupled BPM microservices will skyrocket.
- ProcessMaker 4 – In Q1 ProcessMaker will launch version 4 of its immensely popular ProcessMaker iBPMs platform. Expect true zero code delivery, best of breed open source, 100% event-driven architecture, 100% microservices based, SDKs in all major languages, and a simple, fast, sleek new design.
At Red Hat, we're focused on delivering the next generation of development platforms for modern cloud-native applications, and our BPM and decision management technology is a central part of that strategy. We believe that the business knowledge captured by BPM and decision management systems is critical for the creation of applications that serve the business effectively.
In 2019 I expect to see a convergence of BPM, application development and cloud platforms to support a more diverse community of developers and business users. Most traditional BPM systems are moving in this direction – away from pure process management and towards more comprehensive digital automation, combining BPM with emerging technologies like RPA, AI/ML, low-code, etc. And, of course, Blockchain.
2019 may be the year that we come up with a usable Blockchain-based solution to enhance the integrity of business processes. Most importantly, I think that digital automation will increasingly be delivered from the cloud, via a platform largely built on open source software. I predict that terms like “cloud-native”, “container”, “microservice” and “Kubernetes” will become mainstream and even added to the business analyst’s lexicon over the next year or two!
Decision Management Solutions
Companies will start demanding that their investments in Predictive Analytics, Machine Learning and AI show a real ROI. As a result, those leading the adoption of these technologies will have to stop indulging themselves, stop the science experiments and start delivering value. They’ll find there are two barriers to success – business understanding and the last mile.
Success will require teams to listen to the business not just to the data. They will have to adopt approaches and techniques, like decision modeling and design thinking, to engage business owners in creating shared business understanding BEFORE they start their projects. They will have to kill the “AI program” and create programs focused on business needs.
And once they know what business problem they are trying to solve they will have to invest in getting their analytic or AI solution over the line. They’ll have to invest in the last mile of their solution - business rules, organizational change, decision governance and infrastructure to let the business continually refine and improve the way these technologies are being applied.
2019 needs to be the year AI, ML and Analytics technologists stop SPENDING money and start MAKING some.
Next year we expect to see a new wave of consolidation in the BPM/low/no code/workflow industries where large established and vertically focused software companies will look to add these capabilities to their existing offerings to meet the global demand for digital transformation for their clients. In addition, private equity firms will contribute to the consolidation wave through direct acquisitions and roll up strategies.
Intellect will continue to focus its efforts on developing and providing expertise and solutions for compliance heavy industries, enabled by its no code bpm compliance platform.
Let's be honest - what we expect from the New Year predictions is this: "in year 20xx, the BPM industry will rocket right into the sky" :)
Yet it doesn't happen, year after year. Why? I guess it's because BPM is too strategic: it's a powerful magic indeed, but it's too long-term. It's about controlling the second derivative: instead of cutting costs (affecting the target function directly) or making one-time process improvements (first derivative of the target) we aim to develop sustainable internal capability to improve any process, any time. Seriously? At the end of the day the positive second derivative would win... but only if short-term and mid-term survival is granted, which usually isn't the case.
However, there are some good news: simple recipes of the past don't work well in the Digital era. Continuing business as usual would kill your organization for sure as soon as your industry falls into the "digital vortex".
To be more specific, digital transformation generates well-justified demand for BPM: digital means new business models which in turn requires not just new business processes (it's the first derivative), but also the ability to change business processes as promptly as never in the history (the second derivative).
The balance between short term vs. long term benefits is shifting eastward. RPA is a good illustration for this trend: as Deloitte points out in their recent report, the biggest issue of RPA implementations are complicated and poorly understood business processes that must be streamlined. Humans may be comfortable with implicit processes; robots are not.
Digital Decisioning and DMN will continue to play an essential role in BPM. I can see two major trends in this development:
- Simplification. Representation of decision logic within business processes will be de-facto standardized using mainly simple DMN concepts such as decision tables and avoiding complex programming concepts. The simplified approaches such as “Goal-Oriented Decision Modeling” supported by OpenRules will continue to prevail in development of decision models incorporated into real-world business process models.
- Addressing Complex Decision Optimization Problems. So far, human decision modelers were forced to describe exactly HOW to find a decision by handling all possible combinations of business factors using business rules with multiple exceptions on top of exceptions. More powerful decision engines will allow decision modelers to concentrate on WHAT instead of HOW and will automatically determine multiple feasible decisions and select the optimal decision. I expect that in 2019 a model-based approach to decision modeling (as an alternative to the method-based approach) will start penetrating real-world decisioning applications. The examples of such approach are described here and here. As a vendor of the popular decision management product, we consider this approach as an important differentiator and OpenRules will continue its R&D in this direction.
Over the next 3 years, we will see disruptive change in the architecture of process automation. Here are 5 predictions for intelligent process automation in 2019:
- Tasks will continue to be the primary focus of business process improvement, with integration moving more directly into the process tier while also complementing RPA
- Workflow and forms will subordinate to SaaS applications to provide a seamless inline user experience
- Enterprises will begin to learn what it really means to build a process around a customer. First movers will be profoundly disruptive.
- Fine grained-process metrics will be used more strategically to lay the foundation for IPA prediction machines
- AI will flood the market in 2019, more than tripling the tasks available to automate within a 3-year period
At IDC, we’ve re-structured our process-centric market research around IPA, launched a global initiative around Future of Work, hired a couple of great analysts to expand our coverage, and have been working on our thesis for how both IPA and Future of Work will evolve. Both will be presented at our upcoming Directions event, which is a free, invitation only event. Anyone interested can send me a request.
With the rise of Robotic Process Automation over the past 2 years, we've seen that "process" is hot again. But, with each wave of excitement comes the reality of making these new automation technologies really work and deliver value to an organization.
We predict in 2019, we'll see organizations take a step back and look at holistic approach to Intelligent Automation based on 4 key technologies: Robotic Process Automation, Integration, Artificial Intelligence, and Business Process Management. Together, these form a complete Intelligent Process Automation practice that will help organizations go beyond robotic automation of single tasks, and apply process automation to the full lifecycle of business operations and customer journeys.
2019 will be an important year for BPM as its role as a central engine for digital business will be thrust back into the spotlight.
BPM will be recognized as critical to RPA project success. Publicity-wise, BPM has taken a backseat to emerging areas of business automation technology such as Robotic Process Automation (RPA). However, as the reality sinks in that standalone RPA projects typically do not reach anticipated scale, maturity or impact, the role of business rules and workflow will return to the spotlight. BPM decision makers will favor platforms that seamlessly incorporate robotic work (such as robotic automations and chatbots) alongside human work within workflows. Similarly, RPA decision makers will elevate the role of BPM in their evaluation criteria.
Operational intelligence will make AI more valuable to BPM. Anticipation around AI integration into traditional BPM platforms has created a lot of noise over the past couple of years, but AI has yet to change the BPM game. In 2019, AI will show up in BPM in more tangible ways that will lead directly to real business benefits. For example, by using AI for process discovery or operational intelligence, business leaders can better identify process gaps that cause workplace inefficiencies. AI-enabled data engines will provide insight about what processes can be redesigned and/or automated.
2019 will see BPM practices/technologies deepen the commitment to managing/automating less structured processes as the distinction between process and case settles into a generally understood continuum of activity types rather than hard differences between the two. BPMN will remain the dominant modeling language, but CMMN will continue its incremental adoption as a joint effort by the Department of Veteran Affairs, the Mayo Clinic, and academic institutions showcases its value in health care situations.
Process Modeling tools will continue along some combination of three current trends: extending modeling into Business Architecture territory; presenting better model repository management for analyzing standard normative forms of processes vs. variant production forms of those processes; and offering low-end automation of simple modeled processes. The trend towards seeing BPMS and Case Management System platforms as application development platforms, where process and case modeling are programming motifs, will continue because it enhances the utility of the otherwise staid IDE. Inclusion of decision management technologies will become routine, and DMN will become more ubiquitous in its use.
First-gen RPA for automating algorithmic processing paths through legacy applications will break through, and begin to give way to more cognitively-based ones. I will work to stay ahead of these developments.
Processes have traditionally been the glue that coordinates and orchestrates tasks and resources together in a visible and accountable fashion. This theme will continue, but in new ways and approaches based on the popular drivers of 2019.
- Customer experience will be the major driver of process behavior, so we will see a convergence of customer mapping, human computer interface and visible explainable intelligence working in concert to serve customers, employees and partners in an ever improving fashion. There will be an emergence of emotion detection analysis and appropriate tone driven actions guided by intelligent assistants in 2019.
- In 2019 Automation will continue to advance beyond simple tasks to groups of intelligent and simple tasks dynamically orchestrated across multiple time periods with process from real time and straight through unattended activities to high touch skilled activities that require collaboration, mining assisted, cognitive assisted human decisions and actions over time.
- In 2019, the need for actions at the edge of process reach will push intelligent and dynamic actions in and around mostly IoT enabled work. Goal driven and constraint guided processes of various size and scopes will be acting at the edge in an autonomous fashion thus putting a premium on better forms of decision management.
Miguel Valdés Faura
In 2019, I expect to see the world of BPM to continue its evolution to better integrate the growing robotic workforce and a whole new level of intelligence into the coordination of humans, systems and workflows. Should we call it Intelligent business automation?
Whatever the name is, the future of BPM platforms will be about making sure that processes, applications, robots and systems can reshape and adapt themselves as they run. It will be about delivering insights to business automation participants and assisting optimization specialists with predictions and recommendations on potential improvements. And it will always be about continuously delivering better and more engaging customer experiences..
At Bonitasoft we are convinced that this evolution requires a transformation in the way software is developed and delivered. We believe that BPM platforms are enablers of Iterative and incremental software delivery. We also believe that coding will always be involved in any intelligent business automation implementation and that code it's written by developers so we will continue our commitment to make sure that professional developers and DevOps teams are treated as first-class citizens.
E. Scott Menter
For several years, the shape of the BPM problem space (and therefore, the BPM market) has been hazy, difficult to define. 2019 is the year in which some contours will finally emerge. The term "BPM" will narrow, to be used primarily to refer to top-down, multi-million dollar "big picture" solutions. In contrast, digital process automation (DPA) encompasses the array of broader, business-unit-driven applications that form the bulk of the activity of any digital (or digitally-aspiring) business. Critically, DPA is focused on customer touch points, enabling organizations to leverage the rapid development and enhancement features of the technology to stay competitive in a mercurial market.
Finally, in 2019, the shine will come off the RPA apple, as investors and stakeholders realize that there may be nowhere to go once the existing pent-up demand for the technology has been satisfied. Look for RPA customers to take the next step, turning to DPA to extend task automation into enterprise process automation, on their way to becoming true digital businesses.
First let’s set some context. In 1993 Hammer and Champy published Reengineering the Corporation, which led to the widespread use of Enterprise Resource Planning (ERP) systems that tore down the stove pipes between departments and their operation and data. Unfortunately, once such systems were implemented, after much arduous work, they were “cast in concrete” and difficult to modify.
Then in 2002 Howard Smith co-authored the seminal book, Business Process Management: The Third Wave and introduced the business process management system (BPMS) that was built for change. A decade later, Gartner’s V.P., Jim Sinur introduced the idea of the intelligent BPMS (iBPMS) that enriches the traditional BPMS with additional intelligence, including features like cloud computing, event processing, real-time decision-making, and systems crosslinking, including the Internet of Things. Later, Sinur coauthored Business Process Management: The Next Wave which is all about harnessing complexity (e.g. cloud computing, social, mobile, big data, predictive analytics, and the Internet of Things) with intelligent agents and agent-oriented BPM (aoBPM).
2019, oh my! Machine common sense has long been a critical but missing component of Artificial Intelligence (A.I.). Recent advances in machine learning have created new A.I. capabilities, but machine reasoning across these applications remains narrow and highly specialized. Current machine learning systems must be carefully trained or programmed for every situation. But wait! In September 2018, DARPA announced plans to invest $2 billion in artificial intelligence research in a program called “AI Next.” The Machine Common Sense (MCS) program, where DARPA has teamed up with the Seattle-based Allen Institute for AI (AI2) seeks to address the challenge of articulating and encoding human common-sense reasoning for intelligent machines. The MCS program will aim to create machine common-sense services that can help break down the barrier between the narrowly focused A.I. applications of today and the more general A.I. applications of the future. DARPA is onto something very, very new that could radically change our world forever. Get ready for common-sense BPM (csBPM)!
Dr. Mathias Kirchmer
Here are six key trends and predictions I expect for 2019 and the following years in the field of business process management (BPM).
- Digital Transformation Management: Process management becomes the discipline of digital transformation management, leveraging new technologies as improvement approaches. BPM helps to scope digital transformation initiatives as well as manage the value identification and realization. It defines required transformation projects and appropriate project portfolios. The BPM-Discipline aligns people, products and processes for the digital world.
- Pragmatic Agility for Process Improvements: The volatile business environment requires a fast adaptive approach to improvements and transformation. However, there is also a big need to set clear direction and focus. The combination of agile principles, like the fast realization of process improvements in different stages, with top down approaches and enabling digital tools, such as process mining or prioritization applications, address those challenges. Integrated customer journey planning becomes a major component of process improvements to ensure an outside-in view and the right degree of standardization.
- Value-driven Robotic Process Automation (RPA): RPA has the potential to close the automation gap of traditional applications to deliver significant efficiency gains and other benefits. However, this requires a thought-through process-led approach that considers up and down-stream effects of (ro)bots and realizes their full potential. Organizations recognize this and move towards a systematic approach to realizing value through RPA, leveraging process repositories, mining and other process management tools.
- Business Context for Artificial Intelligence (AI): More and more organizations are excited about the potential opportunities of AI and experiment with topics like Machine Learning (ML) or predictive analytics. In the coming months and years a key focus will be on identifying business scenarios to create best value and strategic advantages through AI. Process-led approaches play a major role.
- Integrated Process and Data Governance: Digital processes are only agile and deliver continued value if they are governed systematically across different departments. The new speed of digital execution accelerates negative effects of bad data quality. Therefore an integrated process and data governance becomes more and more crucial for successful digital processes.
- Hybrid Workforce Management: In the digital enterprise human and digital workforce co-exist. This requires an appropriate management approach to employees who have to resolve more and more often complex exception cases and specific individualized customer requirements. Standard processes are mainly supported through robots - that need to be adjusted and aligned with changing business environments, too. This hybrid workforce management and related tools and techniques play a key role in our digital age.
Fujitsu North America
2019 will be an uneventful year for the business process space. 2018 was clearly a year that highlighted the rise of Robotic Process Automation (RPA) into the public consciousness. RPA fills an important need: it provides a synthetic API for legacy applications that should -- but don't -- have a good web API. In certain cases organizations have to pay people to do simply key entry into these legacy applications. RPA is a powerful tool to automate this, and therefor save a lot of money directly. RPA offers a significant capability In this narrow usage, however the hype in 2018 was that it could do a lot more, and was a general solution for all kinds of business processes. In 2019 RPA will start to enter the trough of disillusionment as people begin to realize that the approach is fragile: it will return a clear value when accessing a legacy application that has been static for many years and can be expected to remain so, however it is less suitable for an application that is actively maintained, and for which the UI is continually changing/improving. The ongoing maintenance code of RPA must be considered. RPA lacks a strong model of human collaboration, and will fail to support general business processes in the same way that BPEL failed to do so 10 years ago. RPA interest will remain strong while the public comes into general understanding of the limitations.
While this is going on, machine learning will deliver unanticipated strong results. We will begin to see the first of the truly "un-programmed" business processes where the system simply learns what to do. People simply forward a request on to other people, and the system learns the processes -- in a way similar to Alpha Zero -- without knowing more than the basic rules. Such machine learning will allow the system to adapt to route work to people with certain specializations without having to program this logic in. This has all been promised as a future feature but machine learning is now advancing faster than we generally realize.
The move to the cloud will accelerate with companies getting quicker to move the liability for data breaches out of their own facilities and into companies that can specialize in combating the increasing thread of cyber crime. In house data centers will remain in only the stanchest of data control domains: government agencies and industries which have strict data handling requirements that are shielded from competition.
Most BPM professionals recognize that business process as a space is robust, secure, and useful. The main challenge to us as a people lies outside the business processes space, and that is being able to sort truth from fiction in social media and other media as well. This is a huge problem that will effect all members of the public, and many professionals will see that they need to help find a solution or (figuratively) die trying. Algorithmically targeted disinformation is an existential threat to our society as we know it, and the information professional is on the front line to find a solution. We still don't know what that solution is yet.