Hyperautomation is a word that’s often heard in companies, especially in IT departments. Sooner or later, you will probably have to deal with this as a system administrator. What is it and above all: what can it do for you?
‘Hyperautomation’ is a concept that is being bandied about quite a bit lately. But is all this hype really justified? Yes, but as with many things that are complex, you have to put in some effort to fully appreciate its power. But the pay-off can be considerable.
So, what exactly is hyperautomation? It comes down to having as many processes as possible within a company done automatically by smart computers and software. That sounds more difficult than it is because there are many tools that an IT department can use to give shape to hyperautomation. Take robotic process automation (RPA), for example. RPA allows you to automate processes that are not very complex, but which take a lot of time if done by a human. Consider, for example, opening emails containing invoices that need to be processed by putting them in the right ledger account. If a large amount of such operations is involved, this can take up quite a lot of time. An RPA process can do this in no time, and almost always without a single error – while a human being can sometimes fail.
RPA relies heavily on the low-code/no-code trend, development environments where you can build apps in no time on the basis of pre-programmed blocks. The idea is that, say, a marketing department that wants an app no longer has to go through developers, leaving them time to streamline the big business processes. The practice is a bit more complicated. Usually, such requests still end up with developers, but the benefit is that with the help of low-code, they can produce something faster that makes marketing happy.
Another aspect of hyperautomation is Everything-as-a-Service, so to speak. It starts with software: a large proportion of companies now use Microsoft 365, a textbook example of SaaS. What this means for the system administrator is simple: no more updates! Platform-as-a-Service is a step further. New user? Just tick a few boxes and the new employee can start using the computer and all the company software they need. IT has never been this scalable. And it also works the other way around: if the company downsizes, subscriptions to services can be cancelled within a month.
AI and machine learning
AI and machine learning play a central role in hyperautomation. A smart logarithm can extract information from a mountain of data that can greatly benefit a company, but there must be time to do something with it. Many processes then go a lot further by allowing the AI to make decisions as well. This is something that needs to be monitored from time to time, because ‘smart’ algorithms often turn out not to be so smart after all, and information bias is always lurking. And before you know it, your clever business process might end up ignoring a specific population group, because the algorithm has simply never been confronted with the characteristics of these people before. The colour of their skin, for example.
We must be honest. If your management comes knocking at your door to talk about hyperautomation it will almost never be about how to make your job easier. It will usually be about cutting costs. Ultimately, most companies end up trying to automate as much as possible to save money. And that in itself is of course not a bad idea.