Thought Leadership
There has been a huge buzz around hyperautomation since the global research and advisory firm Gartner listed it as its number one Strategic Technology Trend for 2020. But what is exactly is hyperautomation and how might it change business intelligence?
According to Gartner, hyperautomation operates on the premise that “all processes that can be automated will be automated” and this will dramatically change how we do work and how we do business. Over the last few years, robotic process automation (RPA) has freed human workers from many repetitive, low-level tasks. However, most businesses have learned the hard way that higher-level business processes must also be automated before they will see any real ROI on their automation efforts. One way to accomplish this is to combine the strengths of machine learning (ML) with RPA and process mining. This empowers organizations to move from automating basic tasks to automating the discovery, improvement, and execution of entire complex business processes from end to end.
Combining machine learning, RPA, and process mining for hyperautomation is a powerful combination. However, automating inefficient and broken business processes will not yield any great advantages. Therefore, broken business processes must first be identified and optimized before they can benefit from hyperautomation. Today’s process for large enterprise business intelligence (BI) is a perfect example of a broken process that must first be fixed before we can benefit from its automation.
While enterprises have invested billions of dollars in BI solutions, their efforts typically fail in the last-mile effort, dead-ending in dashboards that are easy to produce but difficult to decipher and keep track of. As a result, most dashboards go unused, untrusted, or both. At a high level, the business process for BI today typically looks something like this:
- Gather, clean, and prepare data from around the organization
- Generate dozens of dashboards featuring hundreds of metrics
- Hope that the busy decision-makers will:
- Monitor dozens of dashboards
- Identify and track significant changes across hundreds of metrics
- Interpret those changes and understand their impact on the business
- Translate the changes into actions that can benefit the business
- Communicate those actions to the right people in the organization at the right time
- Follow up to see if the actions were taken and if they had the desired effect
Every executive and business decision-maker we have spoken with has told us that this system is broken. Instead of promoting data-driven decision-making and data-driven actions, this system rewards the creation of dashboards. Dashboards have become so easy to produce that more and more people create them, which has ironically added more noise than signal to most organizations’ BI efforts. This is because nobody has the time to monitor and interpret these dashboards 24/7, so they often go unused and are now part of the problem instead of being a part of the solution.
The good news is at Astreya (and before that, at RelayiQ) we have been working for several years on optimizing the process of business intelligence and developing a cutting-edge hyperautomation solution for BI long before Gartner made it popular. To optimize enterprise BI, we completely changed the paradigm — instead of expecting people to monitor and mine dashboards for insights, we use process mining and machine learning-driven thresholds to automate the detection of important changes in core metrics and business processes. Once a change is detected, we enable prescriptive alerts containing the actions that should be taken as a result of the change in the core metrics. Finally, we automate the delivery of those prescriptive actions in near-real-time to those who can act on them. We became experts in hyperautomation for BI before it had a name.
Dashboards will always be useful for analysts who have the time and interest to mine them for insight. But hyperautomation for BI will remove the need for dashboards as an insight delivery mechanism for decision-makers who simply want to know what happened and what to do about it. Decision-makers want to take action, not to spend all of their time solving the many puzzles that dashboards present to them.
How might this look in a real business use case? Let’s look at the example of IT asset management (ITAM) for a large tech organization. Imagine that every piece of IT equipment provisioned within an organization is given a digital twin (a virtual representation of a physical object) upon entering into the supply chain. This digital twin reports back information on the physical piece of equipment throughout its lifecycle in the form of data logs, tracking what is installed, how the equipment is being used, how long it has been in the field, etc. Logs can then be process mined to help identify, monitor, and improve upon business processes. Using machine learning, thresholds can be set to detect outliers in the ITAM business process and help to optimize decisions on when equipment needs to be serviced, refurbished and redistributed, or recycled. Rather than store this information in a passive dashboard or report and hope that someone discovers it, prescriptive actions are automatically sent out to those who need to act on them. For example, by applying machine learning to data across the lifecycle of hundreds or thousands of laptops, you could eventually predict when they will reach end of life and automate the ordering and delivering of replacement laptops — completely eliminating any down time for employees. Likewise, the reverse logistics team could know well ahead of time when a surge of old laptops are coming their way, what the problems are that those laptops are likely to have, and recommendations on whether to refurbish and redistribute or to recycle the equipment.
Hyperautomation is of course not limited to the ITAM use case and can be applied to identify, monitor, and optimize processes across your enterprise — which is precisely why analyst groups like Gartner are so bullish on its transformative potential. If you would like to learn more about how Astreya is starting to leverage hyperautomation at some of the world’s largest tech companies, we’d love to chat and learn more about your use case and how we might be able to help!
Brian Borak
Chief Strategy Officer
