Why Human Interaction Matters in Automation

Automated systems don’t design themselves, and good design doesn’t just happen.

As someone who was born with Design Engineer DNA and learned Design Engineering from mentors who put the first person on the moon, I know the importance of having a human in the loop intimately. For that reason alone, I am a believer in the importance of human interaction within automated processes.

The answer to the question of whether we really need humans to interact with automation (or if that’s just a sign of a poorly designed automation) is emphatically that we really do need to maintain some human interactions within our designs. The recent marketing of automation solutions leads many to believe that the only successful automation is one with no human involvement.

In reality, some of the highest value automations act alongside humans as a digital assistant that not only offloads significant work efforts but also hands intermediate and/or final steps back to a human.

Simply put, as long as people are designing products and outcomes for other real people to use, it’s usually necessary and preferable to have a human in the loop.

Perfect Inputs Rarely Exist

It’s common to hear software engineering firms talk about how their designs can help you get away from needing a human in the loop to deliver your product or services.

The caveat is that this goal misses the larger objective nearly every company has – to provide an exceptional, memorable, and positive experience. It’s as if we have become so mesmerized by the technology that we have forgotten how to thrill – let alone consult early and often in the design – the users of the technology

That’s why human interaction matters in automation. The world is full of processes that require complex and custom inputs. It is rarely the case that all inputs and the algorithms these processes plug into are perfect, and it takes human ingenuity and creativity to find ways to resolve the issues a range of complex inputs create. In addition, many use cases have exception cases or complexities which simply are best served by a human being part of the loop itself.

Think of every time you have found yourself at the end of a company’s customer support decision tree without the answer you are looking for. You will, invariably, wind up talking to a real person via phone or chat to resolve your issue. This is not necessarily an example of design failing to automate well, but proof that useful design will often help funnel complex conditions to a better channel for resolution.

Design Thinking Requires Human Input

At the risk of stating the obvious, one of the goals of Design Thinking is to provide an excellent user experience …  and those experiences are had by people.

As Design Thinkers, we should construct products that provide people with the outcomes they need. If a user is engaging with a product for the first time and they smile, then the design is likely to be right – the millions of iPhone owners out there will know exactly what I am talking about. In fact, it’s critical that designers view that journey through a humanistic (and empathetic) lens to deliver products that really have the wow factor.

Conversely, as a leader, it’s important to understand how every piece of code works, how each line builds into the architecture of the product, and what the person on the other end of that code will experience when it’s run. This holistic approach helps to design for the long-term and deliver a human-driven experience that takes into account both the feelings of the people it’s intended for and how they will use the product.

At the end of the day, if you treat people with dignity and respect, you’ll bring a more empathetic tone to your approach as you help them meet their goals. Good design with appropriate human interactions built-in does just that.

Human Consumers Appreciate Human-Led Design

The end output of automation benefits humans, and the sad truth is that automation too often falls short of human expectations.

One example is the number of “bots” designed to execute a defined set of steps but are delivered with no useful way to learn or manage the value they provide, audit their performance, or even readily modify their code.  Designers often overlook the fact that there are multiple layers of “users,” which can lead them to neglect the management, administrative and compliance needs.

Whenever beginning a new project, it is important to consider what the end outcome(s) need to be and what are the benefits of automating the steps. Doing this right from day one also requires looking upstream and downstream for an “end to end” perspective.

Then, in thinking through how to best leverage algorithms to complete rote tasks and processes, you need to assess how removing humans from these steps might impact the experience. For example, are there aspects of machine-human interaction that would benefit from a more flexible or empathetic approach?

An excellent answer to this question can be found in the recent reliance on “resume bots” to filter job applicants.

There is a consensus within the recruiting sector that these automated processes are having an unintended consequence in the hiring phase, namely weeding out well-qualified people who are outside the norm yet have historically brought unique perspectives and inject creative diversity. The problem is that the non-human and unempathetic bot does not consider them qualified for a role based on its rigid “keyword” criteria.

The design must be developed so that the customers get what they need and feel that the solution is a welcome and good fit. And while this design must be a positive experience for them, it is interesting to consider how many technical developers are enamored by the amount of “tech” that can be delivered. On the flip side, users – aka people! – simply want a solution that is easy to use, accurate and fast: human-led design takes this into account.

Because the world isn’t perfect (and humans certainly fall into that category), automation that creates clear pathways for humans to enter the loop to address exceptions and complexities will win every time.

Humans in the Loop = People

At Apexon, we don’t talk about “headcount” but “heart count.” That is an intentional phrasing choice we have made to help us be mindful of the fact that we are a group made up of individual people with their own thoughts, fears, goals and aspirations.

We try to bring this same concept to our solutions and that means taking a humanistic approach that doesn’t shy away from understanding when a human in the loop may be preferable or necessary.

By putting people’s goals first – which includes the personal goals of our individual contributors, as well as the goals of a client – we enjoy a natural competitive advantage over vendors that simply provide “heads on demand.”  This allows us, as a partner, to build a stronger, motivated, inspired and creative team that’s capable of designing and building better software solutions for the client.

Ultimately, people are always a critical part of both the process and business optimization strategy. And while there will always be discussions about automation taking the place of a physical person, we are actually more likely to be thinking about why and how automation augments the human element, even more so when you are looking to provide the perfect digital and product experience.

Apexon’s digital engineers are adept at not only solving the toughest challenges that our customers can throw at us but also the integration of next-generation solutions into workflows and processes. To find out how we can help you achieve the required digital maturity, please fill out the form below.