Companies that have been looking to implement automation tools prior to COVID-19 are still doing it partially due to the return of investment being quite quick in the space. For a category leader such as UiPath, 2020/Q1 has seen no change. Q2 is expected to be relatively flat due to the high level of uncertainty. Q3 & Q4 have an optimistic outlook (e.g. — BluePrism just raised $124m enhancing the RPA market’s continued strength in the face of an economic downturn).
Clients that have already implemented RPA/Automation solutions and are scaling this throughout their enterprise are doubling down on extending these implementations. Perhaps this is triggered due to the fragility and fear that has been caused by the new shift of forced remote working? On the other hand, organizations that haven’t implemented RPA are the ones that have put new explorations on hold.
Anthony Hsiao from Matterway sees a slow-down/freeze in the RPA market as part of their client base is from the automotive industry. You could best describe the current situation as “hesitant-optimism”. For Q2 the client acquisition pipeline doesn’t get canceled, it rather gets postponed once the high level of uncertainty has fainted. Similar to UiPath’s outlook, Q3 and Q4 are also quite optimistic — due to clients from the automotive or airline industry still continuing conversations (even those that are not costumers yet — this shows optimism about a budget lift-oft especially for “delicious” implementations).
Kulpreet Singh, Managing Director of Test Automation at UiPath stated that COVID19’s impact will heavily affect the spending for “nice-to-have” solutions since most organizations are asking themselves how to survive this crisis, maintain their current client base and strategize on how to make up for the current down- phase?
3 classic RPA/automation use cases we see
COVID19 is accelerating the decision-making process of organizations that have already deployed RPA solutions, but still have a high human workforce. RPA/Automation is only implemented in less complex functions and is lacking expertise in understanding how to apply this across multiple functions.
Large service providers that already have got RPA/automation knowledge and want to take their knowledge to the market via platforms (e.g. IBM offering an automation platform for designing, building, and running intelligent automation services, applications, and digital workers on any cloud, using low-code tools wherever possible).
An interesting use case is a large scale organizational transformation using AI with orchestration and changing the operating model.
Automation use cases
Organizations are making a fundamental change in the “way we are working”. This is not just about how we work, but more importantly on how we are executing our tasks. We often hear complaints from practitioners that only a very small amount of the work can be automized (~10%). Organizations that are quite advanced with RPA are expressing similar concerns. There needs to be a paradigm shift in culture so that we understand how tech can make certain processes more effective and subsequently take the workload off our shoulders. To the worker, the technological core (preferably driven by RPA/Automation) should be a de facto that is unnoticeable and by principal make you more efficient. The technological solution should be the one dictating what can potentially be automated and what can’t. To the practitioner, it should make a difference to what is under the hood, but the user shouldn’t even notice that something is automated.
First Roundtable — COVID-19 update on the intelligent enterprise automation space
There are certain implementations that you need in order to understand something before being able to fix a problem. If you combine all technologies that an organization is using, then you would very quickly see that these technologies are not coming together. Instead, they are all close to each other, partially collaborating, but are not fully intertwined. This is where RPA/Orchestration comes in and is trying to answer the question on how all of these complex pieces can be put together. It is also important to re-architect decades-old systems and modernise them. The “geeks” and “gamers” were the pioneers who said that “if it’s too complicated, then we will make it easier”. We are just now starting to arrive at the stage where the “big checks” are starting to deeply think and strategize on how certain RPA/automation solutions could benefit their organizations. It is expected that by 2024, organizations will lower operational costs by 30% by combining hyper-automation technologies with redesigned operational processes (Source: Gartner).
To date, RPA/Automation solutions are mainly present in the back office and are just slowly moving to the mid/front wing. In reality, there aren’t a lot of people out there that are trying to push the boundaries further than task automation, yet. In theory, this where everyone is trying to get to, so it’s logical that we see companies such as Celonis, UiPath, etc emerge, but this is a space that requires continuous improvement. We are still far away before we can harvest the low hanging fruits from this space.
When automating intelligently, you got options, either change the whole structure and modernize the old systems, or you do small tactical implementations that are designed smartly that allow you to do this en masse. Restructuring these processes, thinking more holistically, applying the systemic change, etc, is what automating intelligently truly means.
We should also note that we are taking baby steps. Space might have been around for the past 20 years, but it’s only got its commercial lens on just about 3 years ago. Would you take advice from someone who has got ~3 years of hands-on business experience when asking how to modernize/restructure decades-old business functions? Often times the answer would be “no”.
It is out of the question that there will be a lot of need for orchestration, stronger analytical power, more complex processes that will be automated, but we also need to look behind RPA and understand what to expect. If we are just at the beginning of the Intelligent Automation market, then in the coming years — we will come up with many functions that we previously believed were not capable of being automated. And just like we’ve seen in various verticals, there will be companies that will have a large share of the market with the reputation for being an orchestration company that can offer you solutions not just horizontally, but also vertically.
A glimpse into how we see the future:
Potentially see automation and AI becoming a de facto standard, meaning, it will be a norm in the business world. Currently, we see it as being a bottom-up adoption, but perhaps we will see a top-down approach in the future.
A convergence across enterprise toolboxes (various technologies). There seems to be a gap which a company will fill with a central lab that is responsible for something beyond RPA. A central stack taking care of every single function?
One point is clear and that is that you can’t run advanced tech on older systems (we still see them everywhere). Cloud adoption etc, is crucial and once this foundation has been standardised, then it becomes interesting to see which other technological areas can be unlocked.
Second Roundtable — Summary of some of the insights gathered from the second virtual roundtable:
10 years ago many organizations would have said that there isn’t enough tech available for solving certain problems. We have arrived at a paradigm shift where there is too much tech available, but with a serious lack of market education. Apart from creating automation tools, there is a big need in educating organizations and guiding them in how to implement the right solution
Organizations should be thinking ahead in terms of “automation-implementation” in order to understand where to start and to define the right sources that are needed
Automation will be driven by “angry young people,” who expect computers and software to make their lives easier, not more complex. Booking events, calls, sending emails, managing contacts, tasks, processes, and more should be easily automatable; young people expect companies to be set up for this — and they will have to be, to hire the best
Joona has given an example on how he sees this in his son who is used to certain tasks being automated in Fortnite (videogame) and expects him to believe that this is a defacto in reality too
Introducing process mining is easy, introducing process automation is easy, but changing people’s behaviors so that they work well with these technologies is by far the hardest part of digital transformation
The biggest challenge is always changing management: How do employees become more enthusiastic about working with bots, even though they want their automated outcomes, they typically aren’t willing to put in the effort to learn how to use them
Democratization may be a core development here that allows the workforce to use automation tools to:
- a) become more productive
- b) resolve new problems
Automation and hyper-automation will change the nature of human resource management; all of the layoffs we see with COVID are “gut instinct” layoffs, and not at all data-driven. If most enterprises had automation solutions in place across their business today, they could downsize and cost-save in a much more efficient and accurate way. Today, this is strictly limited to advanced, forward-thinking technical companies (and even many of these companies are so far behind in their IT infrastructure to be able to implement these technologies and get any value from them.
That’s why today basic RPA is the only successful automation technology out in the market. Companies need to get much better about how they manage their data and implement automation tools in the IT stack if they want to be able to maximize the value of automation
Participants were excited about low-code, but ultimately skeptical that the uptake would be there without giving it enough flexibility and keeping it (almost entirely) no code
Ericsson has got a high demand for low-code but is not capable of fulfilling the demand.
There are only a few companies out there that are able to work in a highly complex environment and manage “reality”
How do we create transparency in what is really happening to ensure we make the right decisions. Fact-based decision making on the full scope rather than just one source, which currently seems to be the most common current approach.