Hyperautomation is coming - but should you believe the hype?

John Appleby Profile picture for user john.appleby August 30, 2022
John Appleby of Avantra, explains hyperautomation without the hype - and the business benefits it can bring to the enterprise.

Process Automation and Workflow Automation Concept - Automating the Flow of Tasks Across Work-related Activities in Accordance with Defined Business Rules © ArtemisDiana - Shutterstock
(© ArtemisDiana - Shutterstock)

Over the last year, we have started to see a growing realization: the robotic process automation that large enterprises have invested in simply can’t scale. It’s not meeting demand or expectation.

So, when Gartner named ‘Hyperautomation’ as one of its top trends for 2022, it was well timed. It gave companies struggling to scale their investment in robotics a jolt – should they try something different?

But I’m not convinced people really know what hyperautomation is, and just see it as hype. While it’s given companies a good reason to pause and reflect on their strategy, it’s thrown up more questions than answers.

Gartner defines hyperautomation as a “business driven, disciplined approach that organizations use to rapidly identify, vet and automate as many business and IT processes as possible.”

When you delve deeper, you come to realize that hyperautomation is best considered as a framework for applying the most suited and mature forms of technology to processes that could, and should be, automated to reduce overheads and improve operational efficiency. It might include robotics, artificial intelligence and machine learning, and even low or no code solutions too.

However, describing hyperautomation as a framework is quite conceptual and I can see some organizations finding it hard to understand just how it will transform their business, and how it will impact the data they manage within it.

When I speak to leaders facing this dilemma, I tend to break it down into two parts; automation, as you might expect, and observability. It’s then that the penny drops.

Observability at the heart of hyperautomation

Observability, I believe, is the key to hyperautomation. One of the easiest ways to explain this is to take the concept of a swimming pool. It’s critical to have the right chemical balance so the water remains clean and safe to swim in. This takes maintenance and is why you’ll see attendants regularly testing the water pH when you’re in the pool at a health club.

It’s considered an important task for health and safety standards. But the downside is that it’s manual. It could, however, be automated using a robot that checks the pH and raises an alert via an app when there’s an imbalance.

Automating a repetitive task in this way is not just useful, it also adds value in terms of ‘observing’ and reporting back on what is happening and needs to be done while other important or urgent tasks are completed.

This form of ‘observability’ is an equivalent of AIOps and acts as a foundation for hyperautomation. It frees up time for people to do other more strategic and high value things, reduces errors and, perhaps most importantly, produces an accurate record of what is happening when. The data generated is valuable and can be manipulated into useful insights that can be acted on. It’s for these reasons companies are more compelled to justify the investment in automation and digital transformation.

What are the use cases in the business world?

Explaining things in this way, helps leaders see the value of introducing AI ops to their business. Many will immediately look at how it can be used by finance to automate reconciliations of purchase orders, improve cash forecasting and in effect create a backbone of automated activities that run the business.

Tasks where items ‘in’ and items ‘out’ must match are also ideal for hyperautomation. Think of the reconciliation that must happen on a trading floor. Do the ‘buys’ match up with the ‘sells’? It’s an extremely important question as it could be worth millions. In the same way, stock of widgets coming into a warehouse and going out for distribution must be closely monitored to ensure profitability and stock and order management. You can’t deliver a good customer experience if you don’t know what you have available to sell.

We’ve worked with companies that to date have managed these sorts of risks by employing people to spot anomalies in the data. While people are good at spotting a mismatch, it’s well understood that manual repetitive tasks are ripe for introducing human error.

Machine learning can however overcome the challenge and create systems that can spot-check faster and with more accuracy.

The importance of check and balances

But, and it’s a substantial but, this utopia only works if there are checks and balances built into the automated process. If we go back to the pool scenario, it would be more useful to the pool manager if the robot did something in response to a chemical imbalance. It could, potentially, add the chemicals.

However, some chemicals take time to dilute. There’s therefore a risk that the robot could persistently think there is an imbalance when there isn’t and keep adding chemicals and create a health hazard, the very thing it is designed to avoid.

It’s a good way to illustrate why it’s important for automated workflows to have checks and balances in place so that risks aren’t introduced. In fact, it’s why the term materials requirement handling ‘nervousness’ exists in the world of ERP.

A laptop manufacturer might decide to automate processes that ensure production meets demand, but if there are fluctuations in demand, it’s possible the system will get nervous and over make in anticipation of higher demand later. Once again, this example points to the need for observability, alerts and check and balances in the workflow that supports the automation.

Take this real-life SAP automation example of The Coop Group. As the second largest cash and carry wholesale supply business in Europe, with a workforce of around 80,000, the Coop Group’s SAP environment is inherently complex. Coop Group found its current monitoring solution was always late in supporting new SAP releases and remained difficult to deploy, requiring imports in the SAP system.

But a major migration is no small task. As the Coop Group was one of the first adopters of HANA, a significant upgrade was needed. Their priorities were fast implementation, improved notification time and daily health checks. But as well as monitoring checks in real time, auto-detection and the ability to prevent failures in advance was essential.

One of the major automation benefits was setting up custom checks. Having alarms and escalation for dedicated checks means the right people always know if there is an issue. Daily custom checks that can flex to fit monitoring needs mean time-saving, agility – and compliance. Rochus Tresch, Coop head of SAP development, observed:

We have quite a lot of custom checks that monitor our complex SAP landscape… While we stress test the system before HANA migration, Avantra gives us useful insights into the source and target systems during the HANA migration, even during the SAP downtime phase.

This example is significant for two reasons:

  1. Every transaction — across 2,000 retail outlets and 124 cash and carry markets in Switzerland and other European countries — is documented in the system in less than 150ms.
  2. Coop Group weren’t looking for technological wizardry. But to keep things running, they needed central monitoring, the ability to add and configure new SAP systems, and do it more efficiently.

Can we call this “hyperautomation”? I’d argue yes, because the Coop Group applied the automation discipline across their landscape. But what we call it isn’t important. What really matters is that the automations from the migration made an operational impact.

Cost v value in determining an ROI

But no matter where your business stands on the hyperautomation curve, there’s absolutely no point in investing in automation if you can’t reduce cost and get a payback. That said, cost shouldn’t necessarily be the only driver either. Frankly, you will never have the energy to explore new product development or expand into new markets if you are wrapped up in manual tasks. The potential revenue and profitability must be factored in when weighing up the investments in automation too.

I find the closed mindset on evaluating ROI simply on cost can be a big barrier to change. To get hyperautomation right you really need to pick off the right tasks to automate and the right technology to do it and have a broader view of the world so you can grasp more opportunities. Only then will you come out on the right side of the hype. Discover more about how hyperautomation can give time back to you and your team.

A grey colored placeholder image