How Automation Can Create A Better Future Of Work.
Design processes that use automation, but keep humans-in-the-loop.
This article by Sagi Eliyahu originally appeared here.
Over the last decade or so, fears about automation and its capacity for upending society have been front and center both in the news and in politics. The concept of safeguarding our incomes against automation’s insatiable appetite for our jobs, for example, has even become a centerpiece of our political discourse.
When people think about automation today, they don’t think of technology that empowers or augments, but rather, technology that displaces — and exacerbates issues of inequality and economic opportunity in the process.
There’s little doubt that inequality of economic opportunity in the labor force will remain a key problem for society to solve in the years ahead. But the extent to which automation is sure to eliminate all of our jobs and throw the economy into disarray is overstated. As Google’s chief economist, Hal Varian, said during a forum on the future of work, sponsored by the Stanford Institute for Human-Centered Artificial Intelligence, “Most jobs are more complex than people realize. Automation doesn’t generally eliminate jobs. Automation generally eliminates dull, tedious, and repetitive tasks. If you remove all the tasks, you remove the job. But that’s rare.”
In other words, automation, though likely to eliminate jobs that consist solely of menial tasks, is perhaps more properly seen as a means of augmenting the modern worker, rather than replacing them. Employed correctly, it has the potential to benefit society. As analysts at McKinsey and Company wrote in a May, 2019 report titled Tech for Good, “The development and adoption of advanced technologies including smart automation and artificial intelligence has the potential not only to raise productivity and GDP growth but also to improve well-being more broadly, including through healthier life and longevity and more leisure.”
Now, that may sound a little rosey, but it’s true that the choice of how we use automation — whether we use it for good, and to create a better future of work for all of us, or purely for profit, and the benefit of a select few — remains ours. It’s also true that automation, far from consecrating our doom, provides us an opportunity. It provides us better opportunities to put people first and to improve life. Automation is a means of augmentation, as well as of liberation. If we use automation correctly, it can make us more powerful, and free us up to spend more time focusing on high-value work — as it once did for farmers by way of the ox, and NASA scientists via the computer. This would empower us not only to produce more value for the companies we run, but to begin to design a future that’s more enjoyable across the board.
None of this will happen on its own, however. Business leaders and those charged with implementing automation technology in all its varied forms today have to take seriously the task of using this technology the right way to create positive change — rather than simply to increase short-term profits.
In my work helping some of the world’s largest enterprises utilize automation software strategically, I’ve identified several strategies for using automation technology to that end.
Here are a few that serve as a great way to start.
Think people-first.
For years now business leaders have been neglecting the relevance of people as an asset in increasing efficiency and sales velocity. Instead, we seek to increase efficiency and sales velocity by replacing people with automation technology.
This is foolish. Automation solutions that seek to either replace people entirely or offload repetitive tasks often make the processes they’re supposed to optimize more rigid. Partly as a result, our employees spend more time compensating for the limitations of our technology solutions than they do benefiting from their functionality. And because our preferred means of problem-solving remains outsourcing problems to enterprise automation apps, the number of those apps we use continues to grow. This means the number of apps we force ourselves to adapt to and change-manage (with only marginal improvement) also continues to grow. The result is a self-reinforcing cycle that hurts both our businesses and the people who power them. It further locks us into a culture and procedural standard that mandates inefficiency, along with overworked, disenchanted employees filling technology gaps.
The truth is, to use new automation technology to create a more holistically efficient operational infrastructure, committing to thinking “people-first” is a necessary first step and a helpful governing principle. Before augmenting a process or system with automation, we should ask ourselves, “How will the infusion of this technology into our infrastructure save employees time and allow them to focus most purposefully on the things they’re best at?”
Determine what solution to use in accordance with your answer. Then, think strategically about how to use that solution to this end of augmenting and optimizing employees — rather than replacing them.
Design processes that use automation, but keep humans-in-the-loop.
One of the chief misconceptions business leaders possess about improving their operational infrastructure is that automation is a silver bullet. While automation is a crucial tool in improving and streamlining your internal processes, automation cannot on its own solve the business process challenges of rigidity and inefficiency from which many companies suffer.
Automation only makes peoples’ lives easier if it’s strategically applied — if it helps people more easily do their work. Automation hastily applied, or applied in the manner most companies apply it today, often does the opposite: if it doesn’t outright replace people (which is what happens when companies try and automate certain processes entirely), it doesn’t meaningfully lessen employees’ workload or make processes more efficient.
Why? For one thing, most enterprise automation technology really isn’t designed for that, so much as it’s designed to automate specific, one-to-one tasks. (Transferring data from one platform to another; extracting data from forms; generating a non-actionable email, etc.) And automated workflows that run independently of people become “black boxes” that people lose oversight of. This makes it much harder to change or adapt if needed.
But more importantly, most key processes shouldn’t be automated end-to-end.
The processes your IT team creates are very likely multifaceted and complex. To run correctly — such that they create value of the sort we’ve been discussing — they require something automation technology can’t provide: nuance. A smarter use of automation to improve process design is to use it not as an end itself, but as a means of augmenting employees’ ability to do their job effectively.
That means keeping a human-in-the-loop.
You can think of a “human-in-the-loop” as a kind of traffic cop, or, better yet, someone who can actually practically be that director of the orchestra mentioned in the introduction.
Automation-augmented processes that run under the oversight of a human-in-the-loop prove more efficient, but also more flexible and adaptive. The presence of a human who can make adjustments to the process if needed, or shepherd certain steps to completion, guarantees that the process never breaks down.
Perhaps that’s the biggest value-add: keeping humans-in-the-loop allows for adaptability. A human-in-the-loop can mitigate edge cases and make crucial judgment calls when needed — something no automation tool would ever reliably be able to do. The tasks you do want to automate are the mundane ones that correspond with that judgement call (i.e. finding the record in the ticketing system, updating the record with details, routing it to the next person, etc.).
Bottom line: automation is powerful, but only when used intelligently.
It is not enough to automate one task or simply automate to replace your people. Automation has the potential to create real efficiencies without introducing more complexity, which is imperative towards optimizing your cross-functional processes in all the ways we’ve discussed so far. It also has the real potential to empower us as humans and improve our lives. But only if we use it the right way.
